Method for sizing a two-phase heat exchanger

Abstract

Method for sizing a two-phase heat exchanger. This method comprises a multi-criteria optimization step (102) which determines values, for each hydraulic diameter Dhi and each surface area Ssi, that decrease a cost function F while respecting a laminar flow constraint in each channel of each pass Pai of the exchanger. The cost function F is defined by the following relationship: F ​​= w1*F1 + w2*F2 + w3*F3, where: - w1, w2, and w3 are weighting coefficients, - F1 is a value representing the total pressure drop of the exchanger, - F2 is a value representing the heat exchange surface area of ​​the exchanger, - F3 is a value representing the total footprint of the exchanger. Fig. 9

Description

Title of the invention: Method for sizing a two-phase heat exchanger

[0001] The invention relates to a method for dimensioning a two-phase heat exchanger. The invention also relates to:

[0002] - a method of exchanging heat with a heat transfer fluid using an exchanger heat sized by implementing the sizing process,

[0003] - a two-phase heat exchanger for implementing the heat exchange method heat with a heat transfer fluid, and

[0004] - an apparatus comprising this two-phase heat exchanger.

[0005] Two-phase heat exchangers are used in many systems to condense or evaporate a heat transfer fluid. For example, application US2014096555A1 describes such a two-phase heat exchanger used to condense ammonia in the gaseous state. This application indicates that the hydraulic diameter of the channels inside which the ammonia circulates decreases as the ammonia changes from its gaseous state to its liquid state. In addition, this application describes that the flow of the fluid in the channels must be turbulent to increase the heat exchange with the external environment. To increase the turbulence, application US2014096555A1 proposes placing pads inside the channels.

[0006] Reducing the hydraulic diameter as the ammonia changes from the gaseous state to the liquid state is advantageous because it reduces the size of the heat exchanger. However, reducing the hydraulic diameter increases the pressure losses and therefore the energy consumption required to circulate the ammonia in the channels of this heat exchanger.

[0007] The invention aims to remedy this drawback by proposing a method for dimensioning a heat exchanger which makes it possible, for equal space requirements, to reduce the pressure losses or, for equal pressure losses, to reduce the size of the heat exchanger.

[0008] The subject of the invention is therefore a method for dimensioning a two-phase heat exchanger comprising:

[0009] - an inlet orifice through which the heat transfer fluid enters in one of the gaseous states and liquid and an outlet through which the heat transfer fluid exits in the other of the gaseous and liquid states;

[0010] - Npa passes Pa,, each pass Pa; comprising Nc; channels, where:

[0011] - the number Npa is the total number of passes Pa; in the exchanger, this number Npa being greater than two and chosen in advance;

[0012] - the index i is an order number included in the interval [1; Npa] which identifies the pass Pa; ;

[0013] - the channels of the same pass Pa; all extend parallel to the same longitudinal direction from a respective inlet to a respective outlet;

[0014] - the length of each channel of the same pass Pa; between its inlet and its outlet is equal to a length L; chosen in advance;

[0015] - between its inlet and its outlet, each channel is fluidically isolated from the other channels of the same pass Pa; ;

[0016] - the hydraulic diameter and the flow section of each channel of the same passes Pa; are constant over the entire length L; ;

[0017] - all the channels of the same pass Pa; are practically identical, that is to say that the hydraulic diameters and the flow section areas of these channels are, respectively, between 0.9*Dh; and l,l*Dh; ​​and between 0.9*Ss; and l,l*Ss;, where Dh; and SS; are, respectively, the average hydraulic diameter and the average flow section area of ​​these practically identical channels;

[0018] - in the same pass Pa;, the channels are grouped within Npl plates PI which each contain Nc^ channels and, within each plate Pl .r the Nc^ channels are arranged next to each other in a transverse direction perpendicular to the longitudinal direction, where the index j identifies the channel plate Pf j among the set of plates in the pass Pa;, Npl; is the known number of plates in the pass Pa; and NCij is the known number of channels in the plate Plij;

[0019] - each plate Plij of each pass Pa; has an exchange surface Sechij with an external environment, this exchange surface Sech^ extending mainly in a plane parallel to the longitudinal direction of the pass Pa; and to the transverse direction of this plate Pl^;

[0020] - the distance E^between two immediately consecutive plates Pl^ and PlIJ+i is su smaller than the hydraulic diameter Dh; ;

[0021] - an inlet manifold Co which fluidly connects the inlets of the channels of the passes Pai to the inlet port, the singular pressure drop APs>0 caused by this inlet collector Co being known;

[0022] - collectors Q which connect in series the different passes Pa; in order of increasing hydraulic diameters when the heat transfer fluid exits in the gaseous state through the outlet orifice and in order of decreasing hydraulic diameters when the heat transfer fluid exits in the liquid state through the outlet orifice, so that the heat transfer fluid can circulate from the inlet orifice to the outlet orifice passing only once through each pass, for this purpose each collector Ci being arranged between the previous pass Pa; and the following pass Pai+i, where the previous pass Pa; and the following pass Pai+i are those located, respectively, immediately in upstream and immediately downstream of this collector C; in the direction of circulation of the heat transfer fluid inside this exchanger, the singular pressure drop APs>i caused by each collector C; being known;

[0023] - a CNpa outlet manifold which fluidly connects the outlets of all channels from the PNpa pass to the outlet orifice, the singular pressure loss APs>Npa caused by this outlet collector CNpa being known,

[0024] this exchanger being designed to operate under the following nominal conditions:

[0025] - the total mass flow rate of the heat transfer fluid inside the exchanger is equal to m t ;

[0026] - the pressure and temperature of the heat transfer fluid at the inlet orifice are equal, respectively, to Pin and Tin;

[0027] - the volume enthalpy of the heat transfer fluid at the inlet orifice is equal to hm>in ;

[0028] - the temperature of the external environment is equal to Text;

[0029] - for each plate Phj, the nominal value of the exchange flow density of heat through the exchange surface Sechij of this plate Phj is equal to qwij;

[0030] in which the dimensioning method comprises:

[0031] 1) an initialization step during which initial values ​​Dhij0 and Ssij0 are chosen for all diameters Dh; and all surfaces Ss; of sections Si, then

[0032] 2) a multi-criteria optimization step which determines new values, for each hydraulic diameter Dh; and each surface Ss;, which decrease a cost function F while respecting a laminar flow constraint in each channel of each pass Pa;,

[0033] where:

[0034] - during this multi-criteria optimization step, the value of the cost function F corresponding to given values ​​of the hydraulic diameters Dh; and the surfaces Ss;, is calculated by performing the following operations:

[0035] 2-1) an estimation operation, for each channel of each pass Pa,, of the characteristics physical characteristics of the operation of the exchanger in steady state, by solving the system of equations comprising the following relations:

[0036] [Math.l] dt ôs

[0037] [Math.2] ôGis) ô / GisP ôpisi, Gis)2 / / u —~ F=---â—' ~ si----;—^-7----—t - pd cri si iq and dslpjaissn ôs 2p i crisiidhi si

[0038] [Math.3] ; . aJsUPeisl I] is +—————— * g nr ' ! ^ ,n put)

[0039] [Math.4] pm [a{s}] = p}[s\[l — a ( si!+ pv[ si! a [ s] )

[0040] [Math.5] • i ri ■ i rn ' ■■ ;s > a i. s i=—-——-------

[0041] [Math.6] G(s) = pm(a(s))u(s)

[0042] [Math 7]

[0043] m(s)= G(s)*Ss(s)

[0044] [Math. 8] „ .[ / Jsi l fjsi . .1 / , . fjsi .. a Fi S ; — ;—r + 2 j :—7--:—r I X ! S ■ I 1 ~ X ; S ! : --:—r X : S ■ ■..... 0;! S; \PviSj pjîs! / pj S !

[0045] [Math.9] . . 64 * V, f J S 1 = —7—;------—7 ii\s^dh[s}

[0046] [Math. 10] 64 * v,. f^s) = ^—---ï— u[ s) * dh. [s i

[0047] [Math. 11] ; . aW *pjs! X i s i - -------- ( 1 — or (s) ) * pz isi+crl sJ * pjsl

[0048] [Math 12]

[0049] mT = m(s)Nc(s)

[0050] [Math 13]

[0051] p(0) = Pin,

[0052] in which:

[0053] - s is the curvilinear abscissa which marks the position of a flow section of a channel relative to the inlet,

[0054] - a(s) is the volume fraction of vapor in the heat transfer fluid in the section flow which is located at the abscissa s,

[0055] - pm(a(s)) is the density of the heat transfer fluid in the flow section which is located at the abscissa s,

[0056] -1 is time,

[0057] - G(s) is the mass flux in the flow section located at the abscissa s,

[0058] - p(s) is the pressure at the flow section located at the abscissa s,

[0059] - Fpm(s) is the coefficient of wall friction in the flow section located at the abscissa s,

[0060] - dh(s) is the hydraulic diameter of the flow section located at the abscissa s, dh(s) being a function which associates with each value of the abscissa s the given value of the hydraulic diameter Dh; corresponding,

[0061] - g is gravity,

[0062] - hm(s) is the enthalpy of the heat transfer fluid in the flow section located at the abscissa s,

[0063] - qw(s) is the function which associates with each value of the abscissa s the value qwjj of the heat flux density at the flow section located at this abscissa s,

[0064] - Pe(s) is the perimeter of the flow section located at the abscissa s, Pe(s) being equal to 4Ss(s) / dh(s),

[0065] - Ss(s) is the surface of the flow section located at the abscissa s, Ss(s) being a function which associates with each value of the abscissa s the given value of the surface of the flow section located at this abscissa s,

[0066] - m(s) is the mass flow rate in the flow section located at the abscissa s,

[0067] - pi(s) and pv(s) are the densities of the heat transfer fluid, respectively, at the liquid state and the gaseous state in the flow section located at the abscissa s,

[0068] - hk and hvs are the saturation enthalpies of the heat transfer fluid, respectively, at the liquid state and the gaseous state, these saturation enthalpies being known constants,

[0069] - u(s) is the speed of the heat transfer fluid in the flow section located at the abscissa s,

[0070] - fi(s) and fv(s) are the Darcy-Weisbach friction coefficients of the fluid ca carrier, respectively, in the liquid state and in the gaseous state in the section flow located at the abscissa s,

[0071] - x(s) is the vapor content of the heat transfer fluid in the flow section located at the abscissa s,

[0072] - Vi and vv are the kinematic viscosities of the heat transfer fluid, respectively, at the liquid state and the gaseous state, these kinematic viscosities being approximated to two known constants independent of the abscissa s,

[0073] - Nc(s) is the number of channels of the pass containing the channel comprising the section flow located at the abscissa s,

[0074] - the variables a(s), pm(a(s)), G(s), p(s), Fpm(s), hm(s), m(s), pi(s), pv(s), u(s), fi(s), fv(s) and x(s) being the unknowns of the system of equations, and

[0075] - the functions dh(s) and Ss(s) being initialized from the values ​​given to the hydraulic diameters Dh; and surfaces Ss; before each iteration of this estimation operation, then

[0076] 2-2) an operation of calculating the value of the cost function F for the values data of the hydraulic diameters Dh; and of the surfaces Ss;, where:

[0077] - the cost function F is defined by the following relation: F= Wi*Fi + w2*F2 + w3*F3,

[0078] - Wi, w2 and w3 are weighting coefficients whose sum is equal to one, the coefficients Wi and w2 each being between 0.3 and 0.45 and coefficient w3 being between 0.1 and 0.4,

[0079] - Fi is a value representative of the total pressure drop of the exchanger calculated using the following relationship:

[0080] [Math. 14] F x = ( A Pr + AF s ) / A

[0081] - APr is the regular pressure drop of the exchanger between its inlet orifice and its outlet orifice, this regular pressure drop APr being calculated using the following relationship:

[0082] [Math. 15] i=l

[0083] where:

[0084] - APri is the regular pressure drop of the pass Pa;, this regular pressure drop APri being calculated using the following relationship:

[0085] [Math. 16]

[0086] where:

[0087] - Dhi is the hydraulic diameter of the channels in the pass Pa,, i.e. equal to dh(s i) with S; a value of the abscissa s which corresponds to a flow section of a channel of the pass Pa,,

[0088] - SS; is the surface area of ​​the flow section Si of a channel of the pass Pa;. that is, equal to Ss(s;),

[0089] - sin>i and soutji are the values ​​of the curvilinear abscissa s corresponding to the positions, respectively, from the inlet and outlet of a channel of the Pa pass;,

[0090] - APS is the singular pressure drop of the exchanger between its inlet orifice and its outlet orifice, this singular pressure loss APS being calculated using the following relationship:

[0091] [Math. 17] 4P,= X dP s , ï = 0

[0092] - F2 is a value representative of the heat exchange surface of the exchanger, calculated using the following relationship:

[0093] [Math. 18] / i=Npa 1=Npli \ 2 *Nc.■; * Dh, * L / 12 * Pech * Lechs \ i=l 1=1 /

[0094] - Pech and Lech are, respectively, the depth and width of a front face of the parallelepiped of smaller volume which entirely contains the exchanger, the front face being the face parallel to the main exchange surface of the plates Plij and the depth Pech being equal to the largest of the elements of the list containing all the products NCij*Dhi for i varying from 1 to Npa and for j varying from 1 to Npl;.

[0095] - F3 is a value representative of the total size of the exchanger calculated at using the following relationship:

[0096] [Math. 19] F 3={Hech * Lech * Pech\ / Vmax

[0097] - Hech is the height of the parallelepiped of smallest volume which entirely contains the interchange,

[0098] - Vmax is a maximum desired volume for the exchanger chosen in advance,

[0099] - the laminar flow stress is that, for any value of the abscissa curvilinear s corresponding to a flow section of a channel, the value Re(s) of the Reynolds number in the flow section located at this abscissa s is less than 2000, the value Re(s) being calculated using the following relation Re(s) = a(s)*u(s)*dh(s) / vv + (la(s))*u(s)*dh(s) / vi,

[0100] 3) a step of verification that the new values ​​of the hydraulic diameters Dh; and surfaces Ss; determined during step 2) correspond to a value of the cost function F sufficiently close to its minimum, where the function F is considered sufficiently close to its minimum when the following condition is satisfied:

[0101] [Math.20] i-^pa ________________________________________ ÿ \[Dh: ~ Dh;, , r+iSs.,, ,r<10 Or

[0102] - Dhit and Ssi>t are the current values ​​of the hydraulic diameters Dh; and of the surfaces SS;, and

[0103] - Dhl ( । and Ssijt_i are the values ​​of the hydraulic diameters Dh; and the surfaces Ss; obtained at the end of the previous iteration of step 2),

[0104] 4) if the cost function F is not sufficiently close to its minimum, the process returns to step 2) and, on the contrary, if the cost function F is sufficiently close to its minimum, the optimal values ​​Dhi>opt and Ssi>opt for each diameter Dh i and each surface Ss; are taken equal to the corresponding current values ​​of the diameters Dh; and the surfaces Ss;.

[0105] Embodiments of this sizing method may include one or more of the following features:

[0106] l)The coefficients Wi and w2 are each between 0.37 and 0.43 and the coefficient w3 is between 0.14 and 0.26.

[0107] 2) The coefficients Wi and w2 are equal to 0.4 and the coefficient w3 is equal to 0.2.

[0108] 3) After obtaining the optimal values ​​Dhi>opt and Ssi>opt for each diameter Dh; and each surface Ss;, the method comprises a step of manufacturing the heat exchanger in which, for each pass Pa; of the manufactured heat exchanger, the hydraulic diameter Dh; and the surface Ss; of the flow section of the practically identical channels of this pass are chosen, respectively, between 0.7*Dhi>opt and l.3*Dhi>opt and between 0.7*Ssi>opt and l.3*Ssi>opt.

[0109] The invention also relates to a method for exchanging heat with a heat transfer fluid, this method comprising during nominal operation:

[0110] - the use of a two-phase heat exchanger comprising:

[0111] - an inlet orifice through which the heat transfer fluid enters in one of the gaseous states and liquid and an outlet through which the heat transfer fluid exits in the other of the gaseous and liquid states,

[0112] - Npa passes Pa,, each pass Pa; comprising Nc; channels, where:

[0113] - the number Npa is the total number of passes Pa; in the exchanger, this number Npa being greater than two,

[0114] - the index i is an order number included in the interval [1; Npa] which identifies the pass Pa;,

[0115] - the channels of the same pass Pa; all extend parallel to the same longitudinal direction from a respective inlet to a respective outlet,

[0116] - the length of each channel of the same pass Pa; between its inlet and its outlet is equal to a length L;,

[0117] - between their input and their output, each channel is fluidically isolated from the others channels of the same pass Pa;,

[0118] - the hydraulic diameter and the flow section of each channel of the same passes Pa; are constant over the entire length L;,

[0119] - all the channels of the same pass Pa; are practically identical, that is to say that the hydraulic diameters and the flow section areas of these channels are, respectively, between 0.9*Dh; and l,l*Dh; ​​and between 0.9*Ss; and l,l*Ss;, where Dh; and SS; are, respectively, the average diameter and the average flow area of ​​these practically identical channels, and

[0120] - in the same pass Pa;, the channels are grouped within Npl; plates Plij which each contain Nc^j channels and, within each plate Plir the Nc^j channels are arranged next to each other in a transverse direction perpendicular to the longitudinal direction, where the index j identifies the channel plate Plij among the set of plates in the pass Pa;, Npl; is the number of plates in the pass Pa; and NCij is the number of channels in the plate Pl^,

[0121] - each plate Pl^ of each pass Pa; has an exchange surface Sechij with an external environment, this exchange surface Sech^ extending mainly in a plane parallel to the longitudinal direction of the pass Pa; and to the transverse direction of this plate Pl^,

[0122] - the distance Eij between two immediately consecutive plates Pl^ and Plij+i is su- smaller than the hydraulic diameter Dh;

[0123] - an inlet manifold Co which fluidly connects the inlets of the channels of the Pai passes to the entrance hole,

[0124] - collectors Q which connect in series the different passes Pa; in order of increasing hydraulic diameters when the heat transfer fluid exits in the gaseous state through the outlet orifice and in order of decreasing hydraulic diameters when the heat transfer fluid exits in the liquid state through the outlet orifice, so that the heat transfer fluid can circulate from the inlet orifice to the outlet orifice by passing only once through each pass, for this purpose each collector Q being arranged between the previous pass Pa; and the following pass Pai+i, where the previous pass Pa; and the following pass Pai+i are those located, respectively, immediately upstream and immediately downstream of this collector C; in the direction of circulation of the heat transfer fluid inside this exchanger,

[0125] - a CNpa outlet manifold which fluidly connects the outlets of all channels from the PNpa pass to the outlet orifice,

[0126] - controlling a pump to circulate the heat transfer fluid in the heat exchanger under the following nominal conditions:

[0127] - the total mass flow rate of the heat transfer fluid inside the exchanger is equal to m T,

[0128] - the pressure and temperature of the heat transfer fluid at the inlet orifice are equal, respectively, to Pin and Tin, and

[0129] - the volume enthalpy of the heat transfer fluid at the inlet orifice is equal to hm, im

[0130] - bringing the Sechij exchange surfaces of the Plij plates of the exchanger into contact with an external medium at a temperature Text, so that, for each plate Plij, the nominal value of the heat exchange flux density across the exchange surface Sech^ of this plate Pl^ is equal to qwjj,

[0131] wherein, when operating under nominal conditions:

[0132] - inside each channel, the flow of the heat transfer fluid is laminar, and

[0133] - for each pass Pa; of the heat exchanger used, the hydraulic diameter Dh; and the surface Ss; of the flow section Si of the channels of this pass are included, respectively, between 0.7*Dhi>opt and l.3*Dhi>opt and between 0.7*Ssi>opt and l.3*Ssi>opt, where the hydraulic diameters Dhi>opt and the surfaces Ssi>opt are those calculated for each pass Pa; by implementing the above sizing method to size an exchanger intended to operate under the same nominal conditions and having the same structure and by choosing:

[0134] - Hech and Lech equal to, respectively, the height and the width of the parallelepiped of smallest volume that entirely contains the heat exchanger, and

[0135] - the volume Vmax equal to the volume of this parallelepiped of smaller volume which completely contains the heat exchanger.

[0136] Embodiments of this heat exchange method may include one or more of the following features:

[0137] 1) For each pass Pa; of the heat exchanger used, the hydraulic diameter Dhi and the surface Ss; of the flow section Si of the channels of this pass are included, respectively, between 0.9*Dhi>opt and l.l*Dhi>opt and between 0.9*Ssi>opt and l.l*Ssi>opt.

[0138] 2) The heat transfer fluid is water.

[0139] 3) When using the heat exchanger, the inlet port is located at- above the outlet orifice and the passes Pa; are arranged one below the other in increasing order of index i so that the heat transfer fluid flows from top to bottom.

[0140] 4) The external environment is the ambient air.

[0141] The invention also relates to a two-phase heat exchanger for implementing the above heat exchange method, this heat exchanger comprising:

[0142] - an inlet orifice through which the heat transfer fluid enters in one of the gaseous states and liquid and an outlet orifice (44) through which the heat transfer fluid exits in the other of the gaseous and liquid states,

[0143] - Npa passes Pa;, each pass Pa; comprising Nc; channels, where:

[0144] - the number Npa is the total number of passes Pa; in the exchanger, this number Npa being greater than two,

[0145] - the index i is an order number included in the interval [1; Npa] which identifies the pass Pa,,

[0146] - the channels of the same pass Pa; all extend parallel to the same longitudinal direction from a respective inlet to a respective outlet,

[0147] - the length of each channel of the same pass Pa; between its inlet and its outlet is equal to a length L;,

[0148] - between their input and their output, each channel is fluidically isolated from the others channels of the same pass Pa,,

[0149] - the hydraulic diameter and the flow section of each channel of the same passes Pa; are constant over the entire length L;,

[0150] - all the channels of the same pass Pa; are practically identical, that is to say that the hydraulic diameters and the flow section areas of these channels are, respectively, between 0.9*Dh; and l,l*Dh; ​​and between 0.9*Ss; and l,l*Ss;, where Dh; and SS; are, respectively, the average hydraulic diameter and the average flow section area of ​​these practically identical channels, and

[0151] - in the same pass Pa;, the channels are grouped within Npl plates Pf j which each contain Nc;j channels and, within each plate P1, the Nc;j channels are arranged next to each other in a transverse direction perpendicular to the longitudinal direction, where the index j identifies the channel plate Plij among the set of plates in the pass Pa;, Npl; is the number of plates in the pass Pa; and NCij is the number of channels in the plate Plij,

[0152] - each plate Plij of each pass Pa; has an exchange surface Sechij with an external environment, this exchange surface Sech^ extending mainly in a plane parallel to the longitudinal direction of the pass Pa; and to the transverse direction of this plate Pl^,

[0153] - the distance Ejj between two immediately consecutive plates Pl^ and Plij+i is su smaller than the hydraulic diameter Dh;,

[0154] - an inlet manifold Co which fluidly connects the inlets of the channels of the Pai passes to the entrance hole,

[0155] - collectors C, which connect in series the different passes Pa; in order of increasing hydraulic diameters when the heat transfer fluid exits in the gaseous state through the outlet orifice and in order of decreasing hydraulic diameters when the heat transfer fluid exits in the liquid state through the outlet orifice, so that the heat transfer fluid can circulate from the inlet orifice to the outlet orifice by passing only once through each pass, for this purpose each collector Ci being arranged between the previous pass Pa; and the following pass Pai+i, where the previous pass Pa; and the following pass Pai+i are those located, respectively, immediately upstream and immediately downstream of this collector C; in the direction of circulation of the heat transfer fluid inside this exchanger,

[0156] - a CNpa outlet manifold which fluidly connects the outlets of all channels from the PNpa pass to the outlet orifice,

[0157] this heat exchanger being suitable and intended to operate under the following nominal conditions:

[0158] - the total mass flow rate of the heat transfer fluid inside the exchanger is equal to m T,

[0159] - the pressure and temperature of the heat transfer fluid at the inlet orifice are equal, respectively, to Pin and Tin,

[0160] - the volume enthalpy of the heat transfer fluid at the inlet orifice is equal to h pf

[0161] - the temperature of the external environment is equal to Text,

[0162] - for each plate, the nominal value of the density of the heat exchange flux at across the exchange surface Sechij of this plate Plij is equal to qw;j,

[0163] in which for each pass Pa; of the heat exchanger used, the hydraulic diameter Dhi and the surface Ss; of the flow section Si of the channels of this pass are respectively between 0.7*Dhi>opt and l.3*Dhi>opt and between 0.7*Ssi>opt and l.3*Ss ijOpt, where the hydraulic diameters Dhi>opt and the surfaces Ssi>opt are those calculated for each pass Pa; by implementing the above sizing method to size an exchanger intended to operate under the same nominal conditions and having the same structure and by choosing:

[0164] - Hech and Lech equal to, respectively, the height and the width of the parallelepiped of smallest volume that entirely contains the heat exchanger, and

[0165] - the volume Vmax equal to the volume of this parallelepiped of smaller volume which completely contains the heat exchanger.

[0166] The embodiments of the exchanger may comprise one or more of the following characteristics:

[0167] 1) The majority of collectors have a single cavity inside which unblocks all the outlets of the channels of the Pa pass; and all the inlets of the channels of the Pai+i pass.

[0168] 2) At least 80% of the channels of the same pass Pa; are identical.

[0169] The invention also relates to an apparatus comprising:

[0170] - the above heat exchanger,

[0171] - a first controllable thruster capable of adjusting the flow rate of the external medium between PI plates i, of a first set of plates, this first set of plates containing only the PI plates inside which the heat transfer fluid is in a two-phase state,

[0172] in which the apparatus comprises:

[0173] - a first pressure sensor and a first temperature sensor, capable of measure, respectively, the pressure Pout and the temperature Tout of the heat transfer fluid at the outlet of the heat exchanger, and

[0174] - an electronic control unit configured to:

[0175] - determine a first saturation temperature of the heat transfer fluid at the level of the outlet port as a function of the measured pressure Pout, then

[0176] - control the first thruster so as to maintain the gap between the first determined saturation temperature and the measured temperature, between predetermined thresholds.

[0177] Embodiments of this apparatus may include the following feature:

[0178] - the heat exchanger comprises a collector Ct upstream of which the fluid ca the carrier fluid circulating in the pass Pat is a single-phase heat transfer fluid and downstream of which the heat transfer fluid becomes a two-phase heat transfer fluid inside the pass Pat+i,

[0179] - the device comprises

[0180] - a second controllable thruster capable of adjusting the flow rate of the external medium between Plij plates of a second set of plates, this second set of plates containing only the Plij plates inside which the heat transfer fluid is in a single-phase state,

[0181] - a second pressure sensor and a second temperature sensor, capable of measuring, respectively, a pressure Pt and a temperature Tt of the heat transfer fluid at the collector Ct of the heat exchanger, and

[0182] - the electronic control unit is configured to:

[0183] - determine a second saturation temperature of the heat transfer fluid at the level of the collector Ct as a function of the measured pressure Pt, then

[0184] - control the second thruster so as to maintain the gap between the second determined saturation temperature and the measured temperature Tt, between predetermined thresholds.

[0185] The invention will be better understood on reading the description which follows, given solely by way of non-limiting example and made with reference to the drawings in which:

[0186] - [Fig.l] is a schematic illustration of the architecture of a device comprising a heat exchanger;

[0187] - Figures 2 and 3 are perspective views of the heat exchanger of the apparatus of [Fig.l];

[0188] - [Fig.4] is a perspective view of a channel plate of the exchanger of heat of figures 2 and 3;

[0189] - Figures 5 to 7 are cross-sectional views of several different plates different used in the exchanger of figures 2 and 3;

[0190] - [Fig.8] is a schematic illustration of an electronic processing unit used to size the exchanger in figures 2 and 3;

[0191] - [Fig.9] is a flowchart of a process for sizing the exchanger figures 2 and 3;

[0192] - [Fig.10] is a flowchart of a method of operation of the apparatus of the [Fig.l].

[0193] In this description, the terminology, conventions and definitions of the terms used in this text are introduced in Chapter I. Then, detailed examples of embodiments are described in Chapter II, with reference to the figures. In Chapter III, variants of these embodiments are presented. Finally, the advantages of the different embodiments are specified in Chapter IV.

[0194] Chapter I: Definitions, terminologies and conventions:

[0195] In the figures, the same references are used to designate the same elements.

[0196] In the remainder of this description, the characteristics and functions well known to those skilled in the art are not described in detail.

[0197] Figures 1 to 7 are oriented relative to an orthogonal XYZ coordinate system, where the X and Y directions are horizontal and the Z direction is vertical. Terms such as "above", "below", "top", "bottom", "upper", "lower" are defined relative to the Z direction.

[0198] The term “very low pressure” means a pressure less than 100 mBar (10 kPa) and, typically, less than 70 mBar (7 kPa) or 50 mBar (5 kPa).

[0199] The symbol “*” denotes scalar multiplication.

[0200] Chapter II: Example of embodiment

[0201] Figures 1 to 3 represent a refrigeration device 2, located in an external environment 4. In this exemplary embodiment, the external environment 4 is the ambient air, characterized by an ambient temperature Text and an ambient pressure Pext. The temperature Text is generally between 10°C and 40°C, while the ambient pressure Pext is generally between 870 hPa and 1050 hPa.

[0202] The device 2 comprises:

[0203] - a two-phase heat exchanger 10;

[0204] - a pump 12, which compresses and circulates a heat transfer fluid inside interchange 10;

[0205] - three thrusters 14 to 16 which ensure the flow of the external medium on interchange 10;

[0206] - a regulator 17;

[0207] - an evaporator 18;

[0208] - an electronic control unit 19.

[0209] Here, the propellers 14 to 16 are fans equipped with blades which rotate around respective horizontal axes 20 to 22. To simplify [Fig.l], these axes 20 to 22 are represented as being parallel to the X direction, but in reality, as indicated in the following figures, the axes 20 to 22 are parallel to the Y direction so as to propel the ambient air between plates Plij of the exchanger 10. Subsequently, the numerical references 14 to 16 are used to designate the fans.

[0210] The rotation speed of the blades of each fan 14 to 16 can be controlled by the unit 19. For this purpose, the fans 14 to 16 are connected to the unit 19.

[0211] The pump 12 makes it possible to circulate a heat transfer fluid inside the exchanger 10. It also makes it possible to compress the heat transfer fluid so that its saturation temperature Tsat becomes higher than the temperature Text. For this purpose, the pump 12 comprises:

[0212] - an inlet 30 at which heat transfer fluid in the gaseous state is sucked in;

[0213] - an outlet 32 ​​at which heat transfer fluid in the gaseous state is discharged.

[0214] The inlet 30 is fluidically connected to an outlet 40 of the evaporator 18. The outlet 32 ​​is fluidically connected to an inlet orifice 42 of the exchanger 10. In this embodiment, the mass flow rate of heat transfer fluid is constant or varies little around a nominal value mT.

[0215] The exchanger 10 condenses, by exchanging calories with the external environment 4, the heat transfer fluid which enters, through the orifice 42, in the gaseous state in order to restore, on an outlet orifice 44, the heat transfer fluid in the liquid state.

[0216] The expander 17 is fluidically connected, on one side to the outlet orifice 44 of the exchanger 10 and on the opposite side, to an inlet 46 of the evaporator 18. The expander makes it possible to drop the pressure of the heat transfer fluid so that its temperature Tsat becomes lower than the temperature Text.

[0217] The evaporator 18 transforms the heat transfer fluid received in the liquid state at its inlet 46 into a gas returned at its outlet 40. For this, typically, the evaporator exchanges calories with an object to be cooled which has not been shown. The temperature To of the object to be cooled is higher than the temperature Tsat of the heat transfer fluid at the level of the evaporator 18.

[0218] Here, the heat transfer fluid circulates in a closed loop. In [Fig.l], the direction of circulation of the heat transfer fluid in the different elements of the device 2 is represented by arrows located inside these elements.

[0219] In this embodiment, the heat transfer fluid is water and the temperature To is less than 100°C or 50°C. Under these conditions, the pressure of the heat transfer fluid at the evaporator is very low so that its saturation temperature Tsat is lower than the temperature To of the object to be cooled. Therefore, to have a significant thermal power, i.e. greater than 1 kW or 5 kW, the volume flow rate of the heat transfer fluid in the exchanger 10 must be high, i.e. typically greater than 500 m3 / h or 1000 m3 / h. For such heat transfer fluid flow rates, it must be ensured that the pressure losses of the exchanger 10 are very low. Until now, the known solution for drastically reducing the pressure losses of the exchanger 10 consists of increasing the surface areas of the flow sections of the heat transfer fluid in the exchanger.However, in the case where the heat transfer fluid is water, this leads to such an increase in the size of the exchanger 10 that it becomes difficult to use in practice. Here, the exchanger 10 is designed to have very low pressure losses, compatible with the use of water as a heat transfer fluid, while having a sufficiently small size for it to be usable in practice.

[0220] For this purpose, the exchanger 10 comprises Npa passes Pa; fluidically connected in series between the orifices 42 and 44. The number Npa is the total number of passes Pa; in the exchanger 10. The number Npa is greater than two. The number Npa is chosen suf sufficiently large to achieve the desired thermal power. For example, here, the number Npa is equal to eight. In this text, the index i is an order number included in the interval [1; Npa] which identifies the pass Pa;. Here, the order numbers i are assigned in ascending order in the direction of circulation of the heat transfer fluid. Thus, the pass Pai is the first pass crossed by the heat transfer fluid which enters through the orifice 42.

[0221] In this embodiment, the passes Pa; are stacked one above the other in the Z direction and the orifices 42 and 44 are located, respectively, at the top and bottom of the exchanger 10. Under these conditions, the pass Pai is the one located at the very top in the exchanger 10 while the pass PaNpa is the one located at the very bottom.

[0222] Each pass Pa; comprises Nc; channels which all extend parallel to the same longitudinal direction from a respective inlet to a respective outlet. The number Nc; is greater than or equal to two and often greater than five or ten. The length of each channel of the same pass Pa; between its inlet and its outlet is equal to a length L;. The length L; is the same for all the channels of this pass Pa;. Moreover, here, the lengths L; of all the passes Pa; are equal. Between its inlet and its outlet, each channel of the pass Pa; is fluidically isolated from the other channels of the same pass Pa;. The hydraulic diameter Dh; and the flow section Si of each channel of the same pass Pa; are constant over the entire length L;. The channels of the same pass Pa; are identical or practically identical to each other.By "substantially identical" is meant that the hydraulic diameter of each of these channels is between 0.9*Dh; and 1.1*Dh; and the surface area of ​​their flow section is between 0.9*Ss; and 1.1*Ss;, where, in this case, Dh; is the average hydraulic diameter of the channels of the pass Pa; and Ss; is the average surface area of ​​the flow section of the channels of the pass Pa;. Preferably, more than 80% or 90% of the channels of the pass Pa; are identical so that their hydraulic diameter and their flow section are each equal to the hydraulic diameter Dh; and to the surface area Ss; of the flow section Si.

[0223] The exchanger 10 also comprises collectors Q for fluidically connecting the passes Pa; to each other and to the orifices 42, 44. More precisely, the exchanger 10 comprises:

[0224] - an inlet manifold Co which fluidly connects the pass Pai to the orifice 42;

[0225] - Q collectors which connect the different passes in series in order of hydraulic diameters Dh; decreasing, so that the heat transfer fluid circulates from the orifice 42 to the orifice 44 passing only once through each pass Pa; ;

[0226] - a CNpa outlet manifold which fluidly connects the PNpa pass to the orifice 44.

[0227] Here, the collector Co comprises a single cavity inside which open orifice 42 and all the inlets of the channels of the Pab pass This cavity Co thus makes it possible to homogenize the pressure and the temperature of the heat transfer fluid which penetrates inside the channels of the Pab pass

[0228] For the indices i included in the interval ] 1, Npa[, each collector C; is arranged between the previous pass Pa; and the following pass Pai+i, where the previous pass Pa; and the following pass Pai+i are those located, respectively, immediately upstream and immediately downstream of this collector C; in the direction of circulation of the heat transfer fluid inside the exchanger 10. In addition, each of these collectors Q comprises a single cavity inside which all the outlets of the channels of the pass Pa; and all the inlets of the channels of the pass Pai+i open. This single cavity of the collector Q thus makes it possible to homogenize the pressure and the temperature of the heat transfer fluid which penetrates inside the channels of the following pass Pai+i.

[0229] Similar to the collector Co, the collector CNpa comprises a single cavity inside which the orifice 44 and all the outlets of the channels of the pass PaNpa open.

[0230] Here, the unique cavities of the collectors Ci are parallelepipedal.

[0231] Stacking the collectors Ci of odd index on top of each other forms a first vertical column 46 ([Fig.2]) while stacking the collectors Ci of even index on top of each other forms a second vertical column 48 ([Fig.2]).

[0232] In this embodiment, the temperature Tin of the heat transfer fluid at the orifice 42 is higher than the temperature Tsat(Pin), where:

[0233] - Pin is the pressure of the heat transfer fluid at the level of the orifice 42;

[0234] - Tsat(Pin) is the saturation temperature of the heat transfer fluid at pressure Pin.

[0235] Under these conditions, the first passes Pa; of the exchanger 10 are used to cool the heat transfer fluid until it reaches a temperature close to its saturation temperature. It is only from the moment when the heat transfer fluid has reached its saturation temperature that it can begin to condense. Thus, in the first passes Pa,, the heat transfer fluid is only in the gaseous state. In other words, in the exchanger 10, there is a collector, denoted Ct, upstream of which the heat transfer fluid which circulates in the pass Pat is only in the gaseous state and downstream of which the heat transfer fluid becomes a two-phase heat transfer fluid inside the pass Pat+i. Here, the passes Pai to Pat form a desuperheater. The passes Pat+i to Pa Npa form the condenser. In the example described, the index t of the collector located between the desuperheater and the condenser is equal to two.

[0236] Here, the passes of the desuperheater are mainly placed opposite the fan 14 to be exposed to the flow of the external medium 4 generated by this fan 14. The passes of the condenser are mainly arranged, for their part, in with respect to the fans 15 and 16 to be mainly exposed to the flow of the external environment 4 generated by these fans 15 and 16.

[0237] The exchanger 10 also comprises:

[0238] - a sensor 50 which measures the temperature Tt of the heat transfer fluid at the level of the Ct collector;

[0239] - a sensor 52 which measures the temperature of the heat transfer fluid at the level of hole 44;

[0240] - a sensor 54 which measures the pressure Pt of the heat transfer fluid at the level of the Ct collector;

[0241] - a sensor 56 which measures the pressure Pout of the heat transfer fluid at the orifice 44.

[0242] The unit 19 controls the rotation speed of the fans 14 to 16 according to the measurements of the sensors 50 to 56 to permanently adapt the air flow which evacuates the energy in contact with the passes Pa;. For this purpose, the unit 19 comprises:

[0243] - a programmable electronic calculator 60, capable of executing recorded instructions registered;

[0244] - a memory 62 which contains the instructions necessary to execute the method beyond [Fig. 10].

[0245] As illustrated in [Fig.3], in the same pass Pa;, the channels are grouped within Npl; plates PIij which each contain Nc^ channels, where the index j identifies the plate Plij among all the plates of the pass Pa;. In [Fig.3], not all the references Plij have been indicated to maintain the readability of this figure. Here, the number Npl; of plates per pass increases as a function of the index i and then decreases. For example, here, the numbers Nph to Npl8 are equal, respectively, to two, two, five, five, ten, ten, eight and eight. In addition, here, all the plates Pl^ of the same pass are identical.

[0246] Inside each plate Pl^, the NcLJ channels are arranged next to each other in a transverse direction parallel to the Y direction.

[0247] Each plate Pl^ has an exchange surface Scch, , with the external environment 4. This exchange surface Sechij extends mainly horizontally. The distance Ejj between two plates Pl^ and Plij+i, immediately consecutive in the Z direction, is greater than the hydraulic diameter Dh; and, preferably, greater than 2*Dh;. The distance EijNpii is the distance between the last plate of the pass Pa; and the first plate of the following pass Pai+i. Preferably, for j ranging from 1 to Npl;, all the distances Ej are equal.

[0248] [Fig.4] represents an example of the realization of a Plij plate in the particular case where this plate comprises ten Caijjk channels, where the index k is the order number of the channel of the Plij plate going in the Y direction. In this example, the diameters hydraulic diameters and the flow sections of the channels Cij>2 to Cij,9 are identical. More precisely, the flow sections of the channels Cij>2 to Cij>9 are rectangular here. The channel Cij.io is the symmetrical of the channel Cijj with respect to a vertical plane parallel to the X direction. The hydraulic diameters and the flow sections of the channels Cij i and Cij io are practically identical to those of the channels C;jj2 to C;jj9. More precisely, the flow section of the channel Cm differs from those of the channels Cii2 to Cii9 M'1 1»J'“ only by the fact that its vertical wall located on the side opposite the channel C;jj2 is curved and not flat.

[0249] The exchange surface Sech^ of the plate P;j is composed of an upper wall Pasupij, a lower wall Painfij and two side walls Palatl | d and Palatijjg. The walls Pasup^ and Painfij extend horizontally. The side walls Palatijjd and Palatij g extend mainly parallel to a vertical plane parallel to the X and Z directions. The thickness ePlij of the plate Plij in the Z direction is small compared to its length L; and compared to its width in the Y direction. Thus, the surfaces of the walls Pasupij and Painfij are ten or twenty times greater than the surfaces of the side walls Palatij d and Palatij g. Therefore, subsequently, for simplicity, the exchange surface Sechij is taken equal to the sum of the surfaces of the walls Pasup^ and Painfj. Typically, the length L; is chosen equal to or less than the width of the air flow generated by one of the fans 14 to 16. For this, here, the length L; is equal to or diameter of the fans 14 to 16.

[0250] Figures 5 to 7 show for illustration purposes the cross sections of the passes, respectively, PaB Pa3 and Pa5.

[0251] [Fig.8] represents an electronic processing unit 60, configured to implement the dimensioning method of [Fig.9]. This unit 60 comprises for this purpose a programmable electronic calculator 62 and a memory 64, containing all the instructions and data necessary to implement the method of [Fig.9].

[0252] The method for dimensioning the exchanger 10 will now be described with reference to [Fig.8]. This dimensioning method makes it possible to find, for each pass, values ​​Dhi>opt and Ssi>opt of the hydraulic diameters and the surfaces of the sections Si. More precisely, the values ​​Dhi>opt and Ssi>opt for each pass Pa,, are those which make it possible, for a given maximum size Vmax of the exchanger 10, to minimize the pressure losses. The size of the exchanger 10 is equal to the volume of the smallest parallelepiped entirely containing the exchanger 10. Here, given that all the plates Plij have the same length L; and that they are all stacked on top of each other in the Z direction, the width Lech of this parallelepiped of smaller volume is equal to the length L; to which are added the thicknesses of columns 46 and 48 in the X direction. The height Hech of this parallel The piped of smaller volume is equal to the height, in the Z direction, of columns 46 and 48. Under these conditions, the maximum space requirement Vmax, expressed in m3, is chosen to be greater than 4*Lech*Hech*ePl;j or 10*Lech*Hech*ePl; j.

[0253] To simplify the determination of these values ​​Dhi>opt and Ssi>opt, the dimensioning method is carried out by making the assumption that all the channels of the same pass Pa; are identical and therefore that the hydraulic diameters and the flow section surfaces of all the channels of the pass Pa; are equal, respectively, to Dh; and Ss; . Indeed, given that the channels of the same pass Pa; are practically all identical, this assumption is acceptable.

[0254] In addition to the maximum volume Vmax of the exchanger 10 to be sized, the following structural characteristics of the exchanger 10 are also fixed beforehand:

[0255] - the number Npa of passes Pa; ;

[0256] - the length L; of the channels of each pass Pa; ;

[0257] - the numbers Npl of plates per pass Pa; ;

[0258] - the thickness ePlij of each plate;

[0259] - the distances between the plates;

[0260] - the number Nc^ of channels in each PI plate;

[0261] - the singular pressure losses of each of the collectors Ci.

[0262] The singular pressure drop of a collector Ci depends essentially on its cross-section in a vertical plane. However, here, the distances Eij between the plates and the thicknesses ePl^ of the plates are fixed. Thus, the cross-section of each collector does not depend on the hydraulic diameters Dh; and the surfaces Ss;. For example, the singular pressure drops of the collectors C; are determined experimentally or from charts. Subsequently, the singular pressure drop of the collector C; is denoted APsi, where the index i varies from zero to Npa.

[0263] Under these conditions, the determination of the values ​​Dhi>opt and Ssi>opt makes it possible to determine the width, in the Y direction, of each channel.

[0264] Furthermore, the nominal operating conditions of the exchanger 10 are known. In particular, the following nominal operating conditions are known:

[0265] - the characteristics of the heat transfer fluid used;

[0266] - the total mass flow rate mT of the heat transfer fluid inside the exchanger 10;

[0267] - the pressure Pin and the temperature Tin of the heat transfer fluid at the orifice entry 42;

[0268] - the volume enthalpy hm>in of the heat transfer fluid at the inlet orifice 42;

[0269] - the temperature Text of the external environment 4;

[0270] - the nominal flow rate of the external medium 4 between the plates Pf j;

[0271] - for each plate Phj, the nominal value qwL| of the exchange flux density of heat by convection through the exchange surface Sech^ of this plate Plij.

[0272] Subsequently, by way of example, the heat transfer fluid is water.

[0273] Typically, the volume enthalpy hmin is calculated from, in particular, the pressure Pin and the temperature Tin.

[0274] In the case where the exchanger 10 is used to condense the heat transfer fluid, the operating conditions are chosen so that the temperature Text is well below the saturation temperature Tsat(Pin) of the heat transfer fluid.

[0275] The nominal flow rate of the external medium 4 between the PILJ plates is for example deduced from the nominal rotation speeds of the fans 14 to 16. The nominal rotation speed of a fan is equal to the average rotation speed over a long period of use of the exchanger 10, that is to say a period greater than 24 h or 48 h.

[0276] The density qwLJ of the heat exchange flow between the heat transfer fluid flowing inside a channel of a plate Plij and the external environment is expressed in W / m2. For a plate Pf j, the density qwL| is equal to the ratio Q; j / Sechij, where:

[0277] - Qij is the heat flux, expressed in W, during the heat transfer between the surface Sechij of this plate Plij and the external environment, and

[0278] - Sechij is the exchange surface, expressed in m2, of the Plij plate.

[0279] Here, the density qw;j is assumed to be constant over the entire surface Sechij and therefore over the entire the length L; of a channel. Indeed, it is assumed that the external air flow is sufficiently large so that the temperature Text of the air in contact with the exchange surface Sechij remains, as a first approximation, constant at all points of this surface Sechij. This assumption is generally reasonable and simplifies the design calculations. Finally, during the phase change of a pure body, the phase change temperature is constant. Therefore, the temperature of the surface Sechij can also be assumed, as a first approximation, to be constant at all points of this surface because the temperature of the surface Sechij is close to the temperature of the heat transfer fluid and therefore, here, to the phase change temperature of water.Thus, with these assumptions, each density qwij is assimilated to a constant which can be calculated from the temperature Text, the nominal speed of the fans 14 to 16 and the thermal characteristics of the material used to make the plates Plij such as its surface thermal resistance.

[0280] The dimensioning method begins with an initialization step 100 during which initial values ​​Dhij0 and Ssij0 are fixed for all the diameters Dh; and all the surfaces Ss; of the sections Si of the passes Pa;.

[0281] Then, during a multi-criteria optimization step 102, the unit 60 determines new values ​​for each hydraulic diameter Dh; and each surface Ss;, which decrease a cost function F while respecting a flow constraint (1) laminar in each channel of each pass Pa;. For this, a minimization algorithm with descending direction of a functional is implemented. For example, it can be the Newton-Raphson algorithm.

[0282] For this, during step 102, the value of the cost function F corresponding to given values ​​of the hydraulic diameters Dh; and the surfaces Ss;, is calculated by executing the following operations:

[0283] - an operation 104 of estimation, for each channel of each pass Pa,, of the characteristics physical characteristics during operation of the exchanger 10 in steady state 10, then

[0284] - an operation 106 of calculating the value of the cost function F for the values data of hydraulic diameters Dh; and surfaces Ss;.

[0285] During operation 104 the following physical characteristics of the exchanger 10, during its operation in steady state, are estimated for each channel of each pass Pa: the volume fraction of vapor a(s), the density of the heat transfer fluid pm(a(s)), the mass flow G(s), the pressure p(s), the coefficient of wall friction Fpm(s), the enthalpy hm(s), the mass flow m(s), the densities pi(s) and pv(s) of the heat transfer fluid, respectively, in the liquid state and in the gaseous state, the velocity u(s) of the heat transfer fluid, the Darcy-Weisbach friction coefficients f(s) and fv(s) of the heat transfer fluid, respectively, in the liquid state and in the gaseous state and the vapor content x(s) of the heat transfer fluid. These physical characteristics vary as the heat transfer fluid condenses in the exchanger 10.Thus, these physical characteristics vary according to a curvilinear abscissa s which identifies a particular flow section crossed by the heat transfer fluid when it flows inside the exchanger 10. Therefore, estimating the preceding physical characteristics consists, for each of these physical characteristics, in determining a function which, to each value of the abscissa s, associates an estimate of the value of this physical characteristic at this location.

[0286] Here, the origin of the curvilinear abscissa s is located at the level of the orifice 42. Thus, each value of the abscissa s corresponds to the position, along the path traveled by the heat transfer fluid, of a flow section of the heat transfer fluid relative to the orifice 42. The maximum value of the abscissa s is reached at the level of the outlet orifice 44. For example, the abscissa s is expressed in meters. The path traveled by the heat transfer fluid between the orifices 42 and 44 is typically greater than several meters. Subsequently, the values ​​sin>i and soutji of the curvilinear abscissa s correspond to the positions, respectively, of the inlet and the outlet of a channel of the pass Pa;.

[0287] To estimate these physical characteristics, a system of equations (A) is solved by making the following assumptions:

[0288] - Hypothesis 1): the channels of the same pass Pa; are all identical, and

[0289] - Hypothesis 2): the heat losses in the collectors C; are negligible.

[0290] Because of hypothesis 1), the physical characteristics of the exchanger are identical in each of the channels of a pass Pa;. Thus, it is sufficient to estimate these physical characteristics in only one of the channels of a pass. Therefore, unless otherwise indicated, the equations of the system of equations (A) are given in the case of a single channel.

[0291] Because of hypothesis 2), the temperature T(sin>i+i) at the inlet of a channel of a pass Pai+i is equal to the temperature T(sout>i) at the outlet of the channel of the previous pass Pa;. In addition, the pressure p(sin>i+i) at the inlet of a channel of a pass Pai+i is equal to p(sout>i) - APsi, where:

[0292] - p(sin>i+i) and p(sout>i) are the pressures at, respectively, the inlet of a channel of the Pai+i pass and the outlet of a channel of the Pa, pass,

[0293] - APs>i is the singular pressure loss of the collector C;.

[0294] Finally, it should be noted that only the physical characteristics of the exchanger at the level of the channels are estimated. These physical characteristics are not estimated for the values ​​of the curvilinear abscissa corresponding to the collectors Ci. Thus, the system of equations (A) is only solved for the values ​​of the abscissa s which correspond to flow sections located inside the channels.

[0295] The system of equations (A) is formed by the union of the following relations (1) to (13).

[0296] Relation (1): Equation of conservation of mass in a channel

[0297] [Math.21] apJtLsH riG s; --------+---- and ds

[0298] where:

[0299] - a(s) is the volume fraction of vapor in the heat transfer fluid in the section flow which is located at the abscissa s, this volume fraction of vapor being defined by relation (5) below,

[0300] - pm(a(s)) is the density of the heat transfer fluid in the flow section which is located at the abscissa s, this density being defined by relation (4) below,

[0301] -1 is time,

[0302] - G(s) is the mass flux in the flow section which is located at the abscissa s, this mass flux being defined by relation (6) below.

[0303] It is emphasized that relation (1) indicates that, in steady state, the mass flux G(s) inside a channel is a constant. Thus, to facilitate the resolution of the system of equations (A), the first term of relation (1) is taken equal to zero.

[0304] Relation (2): Equation of conservation of momentum in a channel

[0305] [Math.22] G is J o I GïS: ï cpisj "s—a— —:—:—a != ' and ES [ prayed;] ] os G h? 2pja\s]\dh[s} f lit ' Or :

[0306] - p(s) is the pressure at the flow section located at the abscissa s,

[0307] - Fpm(s) is the coefficient of wall friction in the flow section located at the abscissa s, this coefficient being defined by the relation (8),

[0308] - dh(s) is the hydraulic diameter of the flow section located at the abscissa s,

[0309] - g is gravity,

[0310] dh(s) is the function which associates with each value of the abscissa s the current value of the corresponding hydraulic diameter Dh;. This function dh(s) is known. Following step 100 and during the first iteration of step 102, the current values ​​are the values ​​Dhij0.

[0311] In steady state, the first term of relation (2) is zero.

[0312] In the case where the plates Pf j extend mainly horizontally, the influence of gravity on the flow of the heat transfer fluid inside a channel is negligible. Thus, to simplify the resolution of the system of equations (A), the last term of relation (2) can be neglected.

[0313] Relation (3): Energy conservation equation, integrated in the steady state with the hypothesis that, within the same pass Pa,, the density qwjj and the mass flow rate are constant over the entire length L; of each of the channels of this pass Pa; and identical in each of the channels of this pass Pa;.

[0314] [Math.23] qjs\*Pe[s\ M * s misi

[0315] where:

[0316] - hm(s) is the enthalpy of the heat transfer fluid in the flow section located at the abscissa s,

[0317] - Pe(s) is the perimeter of the flow section located at the abscissa s, Pe(s) being equal to 4Ss(s) / dh(s),

[0318] - Ss(s) is the surface of the flow section located at the abscissa s, Ss(s) being the known function which associates with each value of the abscissa s the current value of the surface of the flow section located at this abscissa s,

[0319] - m(s) is the mass flow rate in the flow section, assumed to be constant within of a channel but which can vary from one pass to the next, this mass flow rate being defined by the relation (7),

[0320] - qw(s) is the density of the heat flux by convection with the medium exterior, assumed to be constant for each channel of the same pass but which can change from one pass to another.

[0321] The function qw(s) is the known function which, to each value of the abscissa s, associates the value qwLJ of the heat flux density at the flow section located at this abscissa s. Within a channel, it is constant and changes only from one pass to the next.

[0322] During the first iteration of step 102, the function Ss(s) associates with each value of the abscissa s, the corresponding initial value Ssi>0.

[0323] Relation (4): Definition of the density of the heat transfer fluid

[0324] [Math.24] P m (a{s.})=pl{.s)(l--a{s]}+ pv[s.){ais]}

[0325] where pi(s) and pv(s) are the densities of the heat transfer fluid, respectively, in the liquid state and in the gaseous state in the flow section located at the abscissa s.

[0326] Relation (5): Definition of the volume fraction of vapor

[0327] [Math.25] , p^sjihjsi-hp ai Sl = '—— --:----------:—----;

[0328] where hk and hvs are the saturation enthalpies of the heat transfer fluid, respectively, in the liquid state and in the gaseous state. These saturation enthalpies hk and hvs are physical properties of the heat transfer fluid which can be measured or obtained from charts. The enthalpies hk and hvs are therefore considered here as being known constants independent of the abscissa s.

[0329] Relation (6): Definition of mass flow

[0330] [Math.26] G(s) = pm(ct(s))u(s)

[0331] where u(s) is the velocity of the heat transfer fluid in the flow section located at the abscissa s.

[0332] Relation (7): Definition of mass flow rate

[0333] [Math 27]

[0334] m(s)= G(s)*Ss(s)

[0335] Relation (8): Definition of the coefficient Fpm(s)

[0336] [Math.28] J s ! = pm If P- ' Pv[si

[0337] where:

[0338] - fi(s) and fv(s) are the Darcy-Weisbach friction coefficients of the ca- fluid loporteur, respectively, in the liquid state and in the gaseous state in the flow section located at the abscissa s, these coefficients f^s) and fv(s) being defined by the following relations (9) and (10), and

[0339] - x(s) is the vapor content of the heat transfer fluid in the flow section located at the abscissa s, this steam title being defined by relation (11).

[0340] When the flow is single-phase, that is to say when the vapor content x(s) is equal to zero or one, the coefficient Fpm(s) is equal to the value of the parietal friction of the only phase present.

[0341] Relation (9): Definition of the Darcy-Weisbach friction coefficient of the heat transfer fluid in the liquid state

[0342] [Math.29] 64 * V f fA s ! = —- u\Si * ah\Si

[0343] where Vi is the kinematic viscosity of the heat transfer fluid in the liquid state.

[0344] Relation (10): Definition of the Darcy-Weisbach friction coefficient of the heat transfer fluid in the gaseous state

[0345] [Math.30] 64 * v fjs]=—--ï— u[s^dh\sj

[0346] where vv is the kinematic viscosity of the heat transfer fluid in the gaseous state.

[0347] The kinematic viscosities vv and Vi are physical properties which depend essentially on the nature of the heat transfer fluid used. The kinematic viscosity varies little as a function of the abscissa s. Thus, here, the kinematic viscosities vv and Vi are taken equal to constants. These constants are measured experimentally or obtained from charts.

[0348] Relation (11): Definition of steam title

[0349] [Math.31] cdsH pjsj 1 r V •, X > SJ = 7----—----7“---7-----— (1 - cries)) * p-N+cd s)* pj.s!

[0350] Relation (12): Total mass flow rate in a pass

[0351] mT = m(s)Nc(s)

[0352] where Nc(s) is the number of channels of the pass located at the abscissa s.

[0353] Relation (13): Initial pressure

[0354] p(0) = Pin

[0355] The system of equations (A) comprises the following thirteen variables: a(s), pm(a(s)), G(s), p(s), Fpm(s), hm(s), m(s), pi(s), pv(s), u(s), fi(s), fv(s) and x(s). Furthermore, the system of equations (A) comprises thirteen equations. The system of equations (A) is therefore solvable. Thus, during step 102, the system of equations (A) is solved by implementing a known method for numerically solving such systems of equations. For example, the numerical solution method described in the following article can be used to solve the system of equations (A): Medale et al: “A one-di-mensional full-range two-phase model to efficiently compute bifurcation diagrams in sub-cooled boilingflows in vertical heated tube”, Journal of Computational Physics, volume 404, 01 / 03 / 2020.

[0356] During operation 106, unit 60 calculates the cost function F. The cost function F is defined by the following relation: F= wpFi + w2*F2 + w3*F3, where:

[0357] - Fi is a representative value of the pressure drop of the exchanger with the given values ​​of diameters Dh; and surfaces Ss; of flow sections;

[0358] - F2 is a value representative of the heat exchange surface of the exchanger 10 with the given values ​​of the diameters Dh; and the surfaces Ss; of the flow sections;

[0359] - F3 is a value representative of the size of the exchanger 10 with the given values ​​of diameters Dh; and surfaces Ss; of flow sections;

[0360] - Wi, w2 and w3 are predetermined weighting coefficients.

[0361] The sum of the coefficients wb w2 and w3 is equal to one. The coefficients Wi and w2 are each between 0.3 and 0.45 and the coefficient w3 is between 0.1 and 0.4. Preferably, the coefficients Wi and w2 are each between 0.37 and 0.43 and the coefficient w3 is between 0.14 and 0.26. Here, the coefficients Wi and w2 are equal to 0.4 and the coefficient w3 is equal to 0.2.

[0362] The value Fi is calculated using the following relation (15):

[0363] [Math.32] F l =(AP+AP s ) / AP s

[0364] where:

[0365] - APr is the regular pressure drop of the exchanger between the ports 42 and 44, and

[0366] - APS is the singular pressure drop of the exchanger between the orifices 42 and 44.

[0367] The regular pressure drop APr is calculated using the following relation (16):

[0368] [Math.33] 4P,

[0369] where APri is the regular pressure drop of the pass Pa;.

[0370] The regular pressure drop APri is calculated using the following relation (17):

[0371] [Math.34] m / 'F p : i —-------------------- f F {s)ds 2 * / Vc ; * Ss^ * Dh< ç THE * :

[0372] where:

[0373] - Dhi is the hydraulic diameter of the channels in the Pa pass;, i.e. equal to dh(s i) with S; a value of the abscissa s which corresponds to a flow section of a channel of the pass Pa;,

[0374] - SS; is the surface area of ​​the flow section Si of a channel of the pass Pa;. that is, equal to Ss(s;).

[0375] It is emphasized that relation (17) is the same whether the heat transfer fluid flowing in a channel is single-phase or two-phase.

[0376] The singular pressure drop APS is calculated using the following relation (18):

[0377] [Math.35] l=Spa AP = £ AP Si i-ü

[0378] where APsi is the known singular pressure drop of the collector C;.

[0379] The value F2 is calculated using the following relation (19):

[0380] [Math.36] Ne, ,■ * Dh ; * L. A 2 * Pech * Lechï ' 'f ! :

[0381] where Pech and Lech are, respectively, the depth and width of a front face of the parallelepiped of smaller volume which entirely contains the exchanger having the given values ​​Dh; and Ss; of the hydraulic diameters and the surfaces of the flow sections Si. The front face is the face parallel to the main exchange surface of the plates Plij.

[0382] Here, the width Lech is equal to the sum of the width Li and the thicknesses of columns 46 and 48 in the X direction. The depth Pech is equal to the largest of the elements of the list containing all the products NCij*Dhi for i varying from 1 to Npa and for j varying from 1 to Npl;.

[0383] The value F3 is calculated using the following relation (20):

[0384] [Math.37] Pf— <Hech * Lech * Pech\lVmax

[0385] where Hech is equal to the height of column 46 or 48.

[0386] During step 102, the constraint (1) to be satisfied is that in each channel of each pass Pa;, the value Re(s) of the Reynolds number in the flow section located at the abscissa s is less than 2000. Thus, the constraint (1) imposes that the flow of the heat transfer fluid in each of the channels is a laminar flow. For this, the value Re(s) is here calculated using the following relation (21):

[0387] [Math 38]

[0388] Re(s) = a(s)*u(s)*dh(s) / vv + (la(s))*u(s)*dh(s) / vi

[0389] Once the new values ​​of the hydraulic diameters Dh; and the surfaces Ss; have been determined during step 102, during a step 108, the unit 60 verifies that these new values ​​of the hydraulic diameters Dh; and the surfaces Ss; correspond to a value of the cost function F sufficiently close to its minimum. For this, the unit 60 verifies whether the following condition (1) is satisfied:

[0390] [Math.39] X NDhit- ; = 1

[0391] where:

[0392] - Dhit and Ssi>t are the current values ​​of the hydraulic diameters Dh; and of the surfaces SS; ;

[0393] - Dhl ( । and Ssi>t_i are the values ​​of the hydraulic diameters Dh; and the surfaces Ss; obtained at the end of the previous iteration of step 102.

[0394] If condition (1) is satisfied, the method continues with a step 110 during which the optimal values ​​Dhi>opt of the hydraulic diameters of the channels and the optimal values ​​Ssi>opt of the surfaces of the sections Si are taken equal to the current values, respectively, of the diameters Dh; and of the surfaces Ss;. During this step 110, if the optimal values ​​Dhi>opt and Ssi>opt are not suitable, for example because the depth Pech of the exchanger 10 is considered too great, the structure of the exchanger can be modified by modifying, for example, the numbers Npa, Npl;, Nc^, the length L; and / or the distance Ejj then steps 100 to 108 can be repeated.

[0395] If condition (1) is not satisfied, the method returns to step 102 to find new values ​​of the hydraulic diameters Dh; and of the surfaces Ss; which further minimize the cost function F while respecting the constraint (1). Thus, steps 102 and 108 are repeated until values ​​Dhi>opt and Ssi>opt are obtained for each diameter Dh; and for each surface of each section Si for which the value of the function F is sufficiently close to its minimum while satisfying the constraint (1).

[0396] After step 110, the method continues with a step 112 of manufacturing the exchanger 10. For this, for each pass Pa; and for all the channels of this pass Pa;, the hydraulic diameter Dh; and the surface Ss; of the flow section Si of the channels are chosen, respectively, between 0.7*Dhi>opt and 1.3*Dhi>opt and between 0.7*Ssi>opt and 1.3*Ssi>opt and, preferably, respectively, between 0.9*Dhi>opt and 1.1*Dhi>opt and between 0.9*Ssi>opt and 1.1*Ssi>opt. This tolerance margin on the choice of the diameters Dh; and the surfaces Ss; is for example used to choose the Plij plates from plates already available commercially. Thus, in practice, this can substantially simplify the manufacture of the exchanger 10, without substantially degrading its performance. Alternatively, the diameters Dh; and the surfaces Ss; can be chosen equal to the values ​​respectively, Dhi>opt and Ssi>opt.However, in the latter case, the Dhi>opt and Ssi>opt values ​​do not necessarily correspond to the hydraulic diameters and flow section areas of channel plates already on the market. Thus, in this case, Plij plates often have to be made to measure.

[0397] Once the hydraulic diameter Dh; and the surface Ss; have been chosen, the dimensions of the flow section Si corresponding to this hydraulic diameter and this surface Ss;, are determined. In the embodiment described here, the flow section of the channels is rectangular and the height hc; of these channels is fixed by the thickness ePlij of the plate Plij. Under these conditions, determining the dimensions of the section Si consists only of determining the width lc; of this section Si in the direction Y. Here, the width lc; is equal to l^Ss / Dh.-hc,.

[0398] The operation of the apparatus 2 equipped with the exchanger 10, manufactured during step 110, is now described with reference to [Fig. 10]. More precisely, only the operation of the exchanger 10 is described in more detail. The operation of the other components of the apparatus 2 is, for example, conventional.

[0399] During a step 130, the unit 19 controls the pump 12 to circulate the heat transfer fluid in the exchanger 10. More precisely, during step 130, the unit 19 controls the pump 12 to use the exchanger 10 under the same nominal operating conditions as those used to size the exchanger 10.

[0400] In parallel, the unit 19 also controls the fans 14 to 16 to bring the surfaces Scch, , of the PILJ plates into contact with a flow of the external environment 4. To this end, during a step 132, the unit acquires the temperature measurements from the sensors 50 and 52 and the pressure measurements from the sensors 54 and 56.

[0401] Then, during a step 134, the unit 19 controls the speed of the fan 14 to maintain the temperature of the heat transfer fluid at the outlet of the desuperheater slightly above its saturation temperature. Here, for this, from the pressure Pt measured by the sensor 54, the unit 19 determines the saturation temperature Tsat(Pt) corresponding to this pressure Pt. For this purpose, the unit 19 uses a pre-recorded chart which associates with several possible values ​​of the pressure Pt, the corresponding value of the saturation temperature of the heat transfer fluid. Once the temperature Tsat(Pt) has been determined, the unit 19 calculates a difference ATt using the following relationship ATt = Tt-Tsat(Pt ), where Tt is the temperature measured by the sensor 50. Then, the unit 19 compares this difference ATt to a high threshold Sht and to a low threshold Sbt. The Sht and Sbt thresholds are pre-recorded thresholds in memory 62.For example, here, the thresholds Sht and Sbt are equal, respectively, to 0.5 °C and 0.2 °C. If the difference ATt is above the threshold Sht, then the unit 19 controls the fan 14 to increase the rotation speed of its blades and thus increase the air flow passing between the plates of the passes Pai and Pa2. Conversely, if the difference ATt is below the threshold Sbt, then the unit 19 controls the fan 14 to decrease the rotation speed of its blades and thus decrease the air flow passing between the plates of the passes Pai and Pa2. If the difference ATt is between the thresholds Sht and Sbt, then the rotation speed of the fan 14 is kept unchanged.

[0402] In parallel with step 134, during a step 136, the unit 19 controls the speed of the fans 15 and 16 to maintain the temperature Tout of the heat transfer fluid at the outlet of the exchanger 10 slightly below its saturation temperature. Here, for this, from the pressure Pout measured by the sensor 56, the unit 19 determines the saturation temperature Tsat(Pout) corresponding to this pressure Pout. For this purpose, the unit 19 uses the same pre-recorded chart as that used during step 134. Once the temperature Tsat(Pout) determined, the unit 19 calculates a deviation ATout using the following relation ATout = Tout-Tsat(Pout), then compares this deviation ATout to a high threshold Shout and to a low threshold Sbout. The thresholds Shout and Sbout are thresholds pre-recorded in the memory 62. For example, here, the thresholds Shout and Sbout are equal, respectively, to -0.2 °C and -0.5 °C. If the deviation ATout is above the threshold Shout, then the unit 19 controls the fans 15 and 16 to increase the rotation speed of their blades and thus increase the air flow which passes between the plates of the passes Pa3 to and Pa8. Conversely, if the deviation ATout is below the threshold Sbout, then unit 19 controls fans 15 and 16 to reduce the rotation speed of their blades and thus reduce the air flow passing between the plates of passes Pa3 to Pa8.If the difference ATout is between the thresholds Shout and Sbout, then the rotation speeds of the blades of fans 15 and 16 are kept unchanged.

[0403] In this example, the rotation speed of a fan is increased or decreased in steps of 10%.

[0404] Chapter III: Variants:

[0405] Exchanger variants:

[0406] The number Npa of passes may be greater or less than eight. For example, if the thermal power to be exchanged between the external environment and the heat transfer fluid is greater, the number of passes is increased. In this case, the number of passes may be greater than ten or twenty. Conversely, if the thermal power to be exchanged with the external environment is smaller, then the number of passes may be reduced. In the latter case, the number of passes is less than four or three but always greater than or equal to two.

[0407] If the temperature of the heat transfer fluid entering through the inlet orifice 42 is already equal to the saturation temperature Tsat(Pin), the passes corresponding to the desuperheater can be omitted.

[0408] Alternatively, the plates are embedded in a metal foam with a porosity greater than 95%. This makes it possible to homogenize and improve the heat transfer between the external environment and the heat transfer fluid. The presence of such a metal foam very slightly increases the mass of the device and has the advantage, in return, of substantially increasing the heat exchange surface between the exchanger and the external environment at the cost of a minimal increase in the total pressure drop. This therefore makes it possible to improve the thermal performance of the exchanger without substantially increasing its weight.

[0409] Alternatively, not all the plates PI of a pass Pa; are identical to each other. For example, the plates PI i, differ from each other by the number Nc^ of channels they comprise.

[0410] Other embodiments of the channels are possible. For example, the shape of the cross-section of the channels can be circular, oval or any other suitable shape.

[0411] Alternatively, all the channels of the same pass Pa; are identical.

[0412] Other embodiments of the Q collectors are possible. For example, the cavities of the Q collectors may have the shape of a bent cylindrical pipe to fluidly connect the outlets of the channels of the pass Pa; to the inlets of the channels of the pass Pai+i.

[0413] In another variant, the single cavity of each collector Q is replaced by several cavities. Each of these cavities homogenizes the pressure and the temperature of the heat transfer fluid which leaves a respective subgroup of several channels of the pass Pa; before reinjecting the heat transfer fluid thus homogenized into another respective subgroup of channels of the following pass Pai+i. In this case, the number of channels of the pass Pa; which open into a cavity of the collector is equal to or less than the number of channels of the pass Pai+i which open into this same cavity. Inside the collector Ci, each of these cavities is fluidically isolated from the other cavities of this collector by walls impervious to the heat transfer fluid. Preferably, the number of channels in each subgroup of the pass Pa; is greater than one and less than Ne;, where Nc; is the total number of channels in the pass Pa;.

[0414] In the particular case where the number Nc; of channels of the pass Pa; is equal to the number Nci+i of channels of the pass Pai+i, the number of cavities of the collector Ci can be equal to Ne;. In this particular case, each cavity of the collector C; fluidically connects the outlet of a single respective channel of the pass Pa; to the inlet of a single respective channel of the pass Pai+i. Thus, in this particular case, there is no homogenization of the pressure and temperature of the heat transfer fluid before it enters the next pass Pai+i. In this case, typically, only a portion of the collectors Q of the exchanger are configured not to homogenize the pressure and temperature of the heat transfer fluid before it enters the next pass. Preferably, the proportion of collectors C; which do not homogenize the pressure and temperature of the heat transfer fluid before it enters the next pass, is a minority.However, in a less advantageous embodiment, the majority or all of the collectors C; of the exchanger do not homogenize the pressure and temperature of the heat transfer fluid before it enters the next pass.

[0415] Alternatively, the heat transfer fluid does not flow from top to bottom. In this case, gravity is not used as a driving force, which increases the pressure losses slightly.

[0416] Water can be replaced by other heat transfer liquids, such as for example one of the different refrigerant fluids, whether it is a substance pure or a mixture, identified by a numerical designation (ASHRAE notation) starting with the letter R followed by a number. Thus, the heat transfer fluid can be R12 (dichlorodifluoromethane), R717 (ammonia), R410A or others. The heat transfer fluid can also be any gas to be condensed, such as ammonia in the gaseous state or any liquid to be evaporated.

[0417] Other variants:

[0418] The number of fans may be smaller or larger than three. In a simplified embodiment, only fan 14 is omitted or only fans 15 and 16 are omitted. In a very simplified version, all fans are omitted. In the latter case, it is natural convection that circulates the external medium between the plates Pl;j. When natural convection is used to circulate the external medium between the plates, preferably the plates are arranged vertically, i.e. their longitudinal and transverse directions are parallel to a vertical plane.

[0419] The direction of airflow generated by the fans can be reversed.

[0420] The rotation axes of the fans can also be vertical. In this case, the plates are also arranged vertically.

[0421] Other methods of controlling the fans 14 to 16 are possible. In particular, other values ​​are possible for the thresholds Sht, Sbt, Shout, Sbout.

[0422] In a very simplified variant, the speeds of one, several or all of the fans are not controlled as a function of the temperatures and pressures measured by the sensors 50 to 56. For example, the speeds of the fans which are not controlled are constant. In this case, at least some of the sensors 50 to 56 can be omitted. When none of the fans are controlled, the control unit 19 can be omitted or simplified. This simplified variant is for example suitable for the case where the temperature Text does not vary or hardly varies.

[0423] Other propellers, than fans, can be used to circulate the external medium between the PILJ plates. For example, if the external medium is trapped in a sealed circuit, a pump can be used to circulate the air between the Plij plates. The different variants described above in the particular case of fans can be transposed to propellers other than fans.

[0424] The air used to cool the exchanger can be replaced by another gas or a liquid whose temperature is lower than the saturation temperature of the gas to be condensed. The propellant must then be adapted to the external environment to be propelled between the Plij plates.

[0425] The exchanger 10 described herein may be used in any apparatus where it is necessary to condense a heat transfer fluid. Such apparatuses do not necessarily have the primary function of cooling the external environment. For example, the primary function of an apparatus comprising the exchanger 10 may simply be to condense the fluid heat transfer fluid in order to restore it to the liquid state. In this case, once the heat transfer fluid is in the liquid state, it is not re-injected into the exchanger. For example, the outlet orifice 44 is fluidically connected to a container which collects the heat transfer fluid in the liquid state. In such an embodiment, the inlet of the pump is connected to a source of heat transfer fluid in the gaseous state.

[0426] The teachings given here in the particular case where the exchanger is used to condense the heat transfer fluid can be transposed, without effort, to the case where the exchanger is used to vaporize the heat transfer fluid. In this case, the order of the passes Pa; is reversed so that the hydraulic diameter Dh; increases as the heat transfer fluid moves from the orifice 42 to the orifice 44. To use the exchanger to vaporize the heat transfer fluid, the temperature Text must be higher than the temperature Tsat(Pin) of the exchanger.

[0427] Several of the variants described above can be combined in the same embodiment.

[0428] The fact that the majority of collectors Ci comprise a single cavity inside which all the outlets of the channels of the pass Pa; and all the inlets of the channels of the pass Pai+i open can be implemented independently of the method of dimensioning the hydraulic diameters Dh; and the flow sections Si of these channels. In particular, this can be implemented in exchangers for which the hydraulic diameter Dh; and the surface area of ​​the flow section Si of the channels of a pass Pa; are not comprised, respectively, between 0.7*Dhi>opt and 1.3*Dhi>opt and between 0.7*Si>opt and 1.3*Si>opt.

[0429] Chapter III: Advantages of the described embodiments:

[0430] Sizing the hydraulic diameters Dh; and the flow section surfaces Si using the sizing method described above makes it possible to produce a heat exchanger which, for a given size Vmax, has a very low pressure drop. Thus, when using this exchanger, the pressure drops are limited. Limiting the pressure drops makes it possible, in particular, to save energy because the pump used to circulate the heat transfer fluid consumes less electricity for an identical result. Limiting the pressure drops also makes it possible to increase the flow rate of the heat transfer fluid and therefore to work with heat transfer fluids whose saturation pressure, at room temperature, is very low, that is to say, for example, less than 70 mbar (7 kPa).

[0431] It is also emphasized that it is the particular architecture of the exchanger 10 described here that made it possible to design and develop the dimensioning method described. In particular, it is the organization of the channels of the exchanger in successive passes of practically identical channels, which makes it possible to obtain a system of solvable equations for calculating the different physical characteristics of the exchanger 10 during of its operation, from which the pressure loss of the exchanger can be estimated for given hydraulic diameters Dh; and surfaces Ss; flow sections Si.

[0432] Using water as the heat transfer fluid avoids polluting the environment when the heat transfer fluid accidentally escapes into the environment.

[0433] Arranging the exchanger so that, when in use, the heat transfer fluid flows from top to bottom, allows gravity to be used as the driving force to move the fluid. This further limits the pump's electricity consumption.

[0434] The fact that the external environment is the ambient air simplifies the production and use of the exchanger. Indeed, it is then not necessary to enclose the exchanger inside a box which contains an external environment different from the ambient air.

[0435] The fact that each collector Q has a single cavity into which all the channels of the pass Pa; and all the channels of the pass Pai+i open makes it possible to homogenize the pressure and the temperature of the heat transfer fluid before it enters the channels of the following pass Pai+i. Thus, in each pass, the temperature and pressure conditions are substantially homogeneous. Under these conditions, the determined values ​​Dhi>opt and Ssi>opt are optimal for all the channels of the pass Pa;. It is therefore not necessary, in order to substantially improve the performance of the exchanger, to provide that the channels of the pass Pa; furthest from the fans have a hydraulic diameter different from the other channels of this pass Pa;.In other words, thanks to the unique cavity of each collector C;, the performance of the exchanger can be very close to the optimum while using only practically identical channels in each pass Pa; and therefore while remaining simple to manufacture.

[0436] Controlling the fans makes it possible to adapt the operation of the exchanger to variations in the temperature Text and / or to variations in the mass flow rate mT. In addition, maintaining the temperature Tout close to the saturation temperature of the heat transfer fluid at the outlet orifice ensures that the phase change of the heat transfer fluid that occurs in the exchanger is completed at this outlet orifice. This also saves energy by preventing the temperature of the heat transfer fluid discharged at the outlet orifice from being excessively far from the saturation temperature. Indeed, in the case of a condenser, there is no point in overcooling the heat transfer fluid that leaves in the liquid state. Conversely, in the case of an evaporator, there is no point in overheating the heat transfer fluid in the gaseous state.

[0437] Controlling the fans to maintain the temperature Tt close to the saturation temperature Tsat(Pt) of the heat transfer fluid at the collector Ct ensures that the temperature of the heat transfer fluid in the single-phase state is close to its saturation temperature when it enters the collector Ct. This allows for efficient operation of the exchanger in which the phase change of the heat transfer fluid takes place at a temperature close to its saturation temperature. This also saves energy by preventing the temperature of the heat transfer fluid at the collector Ct from being excessively far from the saturation temperature. Indeed, in the case of a condenser, there is no point in dropping the temperature of the heat transfer fluid below its saturation temperature before reaching the collector Ct. Conversely, in the case of an evaporator, there is no point in raising the temperature of the heat transfer fluid in the liquid state well above its saturation temperature before the collector Ct is reached.

Claims

Claims

1. Method for dimensioning a two-phase heat exchanger comprising: - an inlet orifice through which the heat transfer fluid enters in one of the gaseous and liquid states and an outlet orifice through which the heat transfer fluid leaves in the other of the gaseous and liquid states; - Npa passes Pa;, each pass Pa; comprising Nc; channels, where: - the number Npa is the total number of passes Pa; in the exchanger, this number Npa being greater than two and chosen in advance; - the index i is an order number included in the interval [1; Npa] which identifies the pass Pa; ; - the channels of the same pass Pa; all extend parallel to the same longitudinal direction from a respective inlet to a respective outlet; - the length of each channel of the same pass Pa; between its inlet and its outlet is equal to a length L; chosen in advance; - between its inlet and its outlet, each channel is fluidically isolated from the other channels of the same pass Pa; ; - the hydraulic diameter and the flow section of each channel of the same pass Pa; are constant over the entire length L; ; - all the channels of the same pass Pa; are practically identical, that is to say that the hydraulic diameters and the surfaces of the flow sections of these channels are included, respectively, between 0.9*Dh; and l,l*Dhi and between 0.9*Ss; and l,l*Ss;, where Dh; and Ss; are, respectively, the average hydraulic diameter and the average surface of the flow section of these practically identical channels; - in the same pass Pa;, the channels are grouped within Npl plates PI which each contain NcLJ channels and, inside each plate Pl^, the Nc^ channels are arranged next to each other in a transverse direction perpendicular to the longitudinal direction, where the index j identifies the channel plate Plij among all the plates of the pass Pa,, Npl; is the known number of plates of the pass Pa; and Nc^j is the known number of channels in the plate Plij; - each plate Plij of each pass Pa; presents an exchange surface Sechij with an external medium, this exchange surface Scch, , extending mainly in a plane parallel to the longitudinal direction tudinal of the pass Pa; and to the transverse direction of this plate PILJ; - the distance Eij between two immediately consecutive plates Plij and Plij+i is greater than the hydraulic diameter Dh; ; - an inlet manifold Co which fluidically connects the inlets of the channels of the pass Pai to the inlet orifice, the singular pressure loss APs0 caused by this inlet manifold Co being known; - collectors Q which connect in series the different passes Pa; in order of increasing hydraulic diameters when the heat transfer fluid leaves in the gaseous state through the outlet orifice and in order of decreasing hydraulic diameters when the heat transfer fluid leaves in the liquid state through the outlet orifice, so that the heat transfer fluid can circulate from the inlet orifice to the outlet orifice by passing only once through each pass, for this purpose each collector Q being arranged between the previous pass Pa; and the following pass Pai+i, where the previous pass Pa; and the following pass Pai+i are those located, respectively, immediately upstream and immediately downstream of this collector Ci in the direction of circulation of the heat transfer fluid inside this exchanger, the singular pressure drop APsi caused by each collector Ci being known; - an outlet manifold CNpa which fluidically connects the outlets of all the channels of the pass PNpa to the outlet orifice, the singular pressure drop APs>Npa caused by this outlet manifold CNpa being known, this exchanger being designed to operate under the following nominal conditions: - the total mass flow rate of the heat transfer fluid inside the exchanger is equal to mT; - the pressure and temperature of the heat transfer fluid at the inlet are equal, respectively, to Pin and Tin; - the volume enthalpy of the heat transfer fluid at the inlet orifice is equal to hm>in; - the temperature of the external environment is equal to Text; - for each plate Pl^, the nominal value of the density of the heat exchange flux across the exchange surface Scch, , of this plate Plij is equal to qwij; characterized in that the dimensioning method comprises: 1) an initialization step (100) during which initial values Dhi o and Ssij0 are chosen for all the diameters Dh; and all the surfaces Ss; of the sections Si, then 2) a multi-criteria optimization step (102) which determines new values, for each hydraulic diameter Dh; and each surface Ss;, which decrease a cost function F while respecting a laminar flow constraint in each channel of each pass Pa,, where: - during this multi-criteria optimization step, the value of the cost function F corresponding to given values of the hydraulic diameters Dh; and the surfaces Ss;, is calculated by performing the following operations: 2-1) an operation (104) of estimation, for each channel of each pass Pa,, of the physical characteristics of the operation of the exchanger in steady state, by solving the system of equations comprising the following relations: [Math.40] ôpjdsn acU; A ---"---- +---- = 0 dt ds [Math.41] GP'S ds Ois; - 2—;.......... 2p..as:\dn.s -pm\a\s}]g [Math.42] [Math.43] p m (a(s)) = p / is H 1 - a! s) )+p v î [Math.44] ; , pjspjsj-hj £X i S i = ;— —----—-----; “T”” ....... p J s )\h ir [ s H hj+p, J shh P} [ s ; - h,J [Math.45] G(s) - pm(a(s))u(s) [Math 46] m(s)= G(s)*Ss(s) [Math.47] fJS? I ; . , VT / JS! , ,3 XîSl h --Xis r P: * ! ' pjs; [Math.48] 64 * v > f,isl=—--— ul sî * dhis) [Math.49] 64 * v, —-77-— U\ S; * dh\S ; [Math.5O] ( ; have if * p Js ) X i S ; = --------— ;--—— -----;—7 [1 — aisjî * pJsi+cH si * p v !if [Math 51] mT = m(s)Nc(s) [Math 52] p(0) = Pin, in which: - s is the curvilinear abscissa which marks the position of a flow section of a channel relative to the inlet orifice, - a(s) is the volume fraction of vapor in the heat transfer fluid in the flow section located at the abscissa s, - pm(a(s)) is the density of the heat transfer fluid in the flow section located at the abscissa s, -1 is time, - G(s) is the mass flux in the flow section located at the abscissa s, - p(s) is the pressure at the flow section located at the abscissa s, - Fpm(s) is the coefficient of parietal friction in the flow section located at the abscissa s, - dh(s) is the hydraulic diameter of the flow section located at the abscissa s, dh(s) being a function which associates with each value of the abscissa s the given value of the corresponding hydraulic diameter Dh; - g is gravity, - hm(s) is the enthalpy of the heat transfer fluid in the flow section located at the abscissa s, - qw(s) is the function which associates with each value of the abscissa s the value qwjj of the heat flux density at the flow section located at this abscissa s, - Pe(s) is the perimeter of the flow section located at the abscissa s, Pe(s) being equal to 4Ss(s) / dh(s), - Ss(s) is the surface area of the flow section located at the abscissa s, Ss(s) being a function which associates with each value of the abscissa s the given value of the surface area of the flow section located at this abscissa s, - m(s) is the mass flow rate in the flow section located at the abscissa s, - pi(s) and pv(s) are the densities of the heat transfer fluid, respectively, in the liquid state and in the gaseous state in the flow section located at the abscissa s, - hk and hvs are the saturation enthalpies of the heat transfer fluid, respectively, in the liquid state and in the gaseous state, these saturation enthalpies being known constants, - u(s) is the velocity of the heat transfer fluid in the flow section located at the abscissa s, - fi(s) and fv(s) are the Darcy-Weisbach friction coefficients of the heat transfer fluid, respectively, in the liquid state and in the gaseous state in the flow section located at the abscissa s, - x(s) is the vapor content of the heat transfer fluid in the flow section located at the abscissa s, - Vi and vv are the kinematic viscosities of the heat transfer fluid, respectively, in the liquid state and in the gaseous state, these kinematic viscosities being approximated to two known constants independent of the abscissa s, - Nc(s) is the number of channels in the pass containing the channel with the flow section located at the abscissa s, - the variables a(s), pm(a(s)), G(s), p(s), Fpm(s), hm(s), m(s), pi(s), pv(s), u(s), fi(s), fv(s) and x(s) being the thirteen unknowns of the system of equations, And - the functions dh(s) and Ss(s) being initialized from the values given to the hydraulic diameters Dh; and to the surfaces Ss; before each iteration of this estimation operation, then 2-2) an operation (106) of calculating the value of the cost function F for the given values of the hydraulic diameters Dh; and the surfaces SS;, where: - the cost function F is defined by the following relation: F= Wi*Fi + w 2*F2 + w3*F3, - Wi, w2 and w3 are weighting coefficients whose sum is equal to one, the coefficients Wi and w2 each being between 0.3 and 0.45 and the coefficient w3 being between 0.1 and 0.4, - Fi is a representative value of the total pressure drop of the exchanger calculated using the following relationship: [Math.53] F, — ( ZI P F -t- zi PJ / AP s - APr is the regular pressure drop of the exchanger between its inlet and outlet ports, this regular pressure drop APr being calculated using the following relationship: [Math.54] Or : - APri is the regular pressure drop of the pass Pa;, this regular pressure drop APri being calculated using the following relation: [Math.55] 2 * * Ssf * Dh f FpJs)ds Or : - Dh; is the hydraulic diameter of the channels in the pass Pa;, that is to say equal to dh(s;) with s; a value of the abscissa s which corresponds to a flow section of a channel of the pass Pa,, - SS; is the surface of the flow section Si of a channel of the pass Pa;. that is to say equal to Ss(Sj), - sin>i and sout>i are the values of the curvilinear abscissa s corresponding to the positions, respectively, of the inlet and outlet of a channel of the pass Pa,, - APS is the singular pressure drop of the exchanger between its inlet orifice and its outlet orifice, this singular pressure drop APS being calculated using the following relationship: [Math.56] i-Nps AP.- £ APS . - F2 is a value representative of the heat exchange surface of the exchanger, calculated using the following relationship: [Math.57] (i=Npa \ XX Lt / 2* Pech* Lech] ;=1 j = l / - Pech and Lech are, respectively, the depth and width of a front face of the parallelepiped of smallest volume which entirely contains the exchanger, the front face being the face parallel to the main exchange surface of the plates Plij and the depth Pech being equal to the largest of the elements of the list containing all the products NCij*Dhi for i varying from 1 to Npa and for j varying from 1 to Npl;. - F3 is a value representative of the total size of the exchanger calculated using the following relationship: [Math.58] F^iHech * Lech * Pech) / Vmax - Hech is the height of the smallest parallelepiped which entirely contains the exchanger, - Vmax is a maximum desired volume for the exchanger chosen in advance, - the laminar flow stress is that, for any value of the curvilinear abscissa s corresponding to a flow section of a channel, the value Re(s) of the Reynolds number in the flow section located at this abscissa s is less than 2000, the value Re(s) being calculated using the following relation Re(s) = a(s)*u(s)*dh(s) / vv + (la(s))*u(s)*dh(s) / vb 3) a step (108) of verification that the new values of the hydraulic diameters Dh; and of the surfaces Ss; determined during step 2) correspond to a value of the cost function F sufficiently close to its minimum, where the function F is considered sufficiently close to its minimum when the following condition is satisfied: [Math.59] i=Npô _________________________________________ pDh:.-Dhit_JpSs;t-Ss; J where - Dh; t and Ssijt are the current values of the hydraulic diameters Dh; and the surfaces Ss;, and - Dhi>t_i and Ssijt_i are the values of the hydraulic diameters Dh; and the surfaces Ss; obtained at the end of the previous iteration of step 2), 4) if the cost function F is not sufficiently close to its minimum, the process returns to step 2) and, on the contrary, if the cost function F is sufficiently close to its minimum, the optimal values Dhi>opt and Ssi>opt for each diameter Dh; and each surface Ss; are taken equal to the corresponding current values of the diameters Dhi and the surfaces Ss;.

2. The method of claim 1, wherein the coefficients Wi and w2 are each between 0.37 and 0.43 and the coefficient w3 is between 0.14 and 0.

26.

3. The method of claim 2, wherein the coefficients Wi and w2 are equal to 0.4 and the coefficient w3 is equal to 0.

2.

4. Method according to any one of the preceding claims, wherein, after obtaining the optimal values Dhi>opt and Ssi>opt for each diameter Dh; and each surface Ss;, the method comprises a step (112) of manufacturing the heat exchanger in which, for each pass Pa; of the manufactured heat exchanger, the hydraulic diameter Dhi and the surface Ss; of the flow section of the practically identical channels of this pass are chosen, respectively, between 0.7*Dhi>opt and 1.3*Dhi>opt and between 0.7*Ssi>opt and 1.3*Ssi>opt.

5. Method for exchanging heat with a heat transfer fluid, this method comprising during nominal operation: - the use (130) of a two-phase heat exchanger comprising: - an inlet through which the heat transfer fluid enters in one of the gaseous and liquid states and an outlet through which the heat transfer fluid leaves in the other of the gaseous and liquid states, - Npa passes Pa,, each pass Pa; comprising Nc; channels, where: - the number Npa is the total number of passes Pa; in the exchanger, this number Npa being greater than two, - the index i is an order number included in the interval [1; Npa] which identifies the pass Pa,, - the channels of the same pass Pa; all extend parallel to the same longitudinal direction from a respective inlet to a respective outlet, - the length of each channel of the same pass Pa; between its inlet and its outlet is equal to a length L;, - between their inlet and their outlet, each channel is fluidically isolated from the other channels of the same pass Pa;, - the hydraulic diameter and the flow section of each channel of the same pass Pa; are constant over the entire length L;, - all the channels of the same pass Pa; are practically identical, that is to say that the hydraulic diameters and the surfaces of the flow sections of these channels are, respectively, between 0.9*Dh; and l,l*Dhj and between 0.9*Ss; and l,l*Ss;, where Dh; and Ss; are, respectively, the average diameter and the average flow surface of these practically identical channels, and - in the same pass Pa,, the channels are grouped within Npl; plates Plij which each contain NcLJ channels and, inside each plate Plij, the Nc^j channels are arranged next to each other in a transverse direction perpendicular to the longitudinal direction, where the index j identifies the channel plate Plij among all the plates of the pass Pa,, Npl; is the number of plates of the pass Pa; and Nc^j is the number of channels in the plate PI,,, - each plate Pl^ of each pass Pa; has an exchange surface Sechij with an external medium, this exchange surface Scch,, extending mainly in a plane parallel to the longitudinal direction of the pass Pa; and to the transverse direction of this plate Pl .r - the distance E;,j between two immediately consecutive plates Pl^ and Pfj+i is greater than the hydraulic diameter Dh; - an inlet manifold Co which fluidically connects the inlets of the channels of the Pai pass to the inlet orifice, - collectors Q which connect in series the different passes Pa; in order of increasing hydraulic diameters when the heat transfer fluid leaves in the gaseous state through the outlet orifice and in order of decreasing hydraulic diameters when the heat transfer fluid leaves in the liquid state through the outlet orifice, so that the heat transfer fluid can circulate from the inlet orifice to the outlet orifice by passing only once through each pass, for this purpose each collector Q being arranged between the previous pass Pa; and the following pass Pai+i, where the previous pass Pa; and the following pass Pai+i are those located, respectively, immediately upstream and immediately downstream of this collector C; in the direction of circulation of the heat transfer fluid inside this exchanger, - a CNpa outlet manifold which fluidically connects the outlets of all the channels of the PNpa pass to the outlet orifice, - the control (130) of a pump to circulate the heat transfer fluid in the heat exchanger under the following nominal conditions: - the total mass flow rate of the heat transfer fluid inside the exchanger is equal to mT, - the pressure and temperature of the heat transfer fluid at the inlet are equal, respectively, to Pin and Tin, and - the volume enthalpy of the heat transfer fluid at the inlet is equal to hm>in, - bringing the exchange surfaces Sechij of the plates Plij of the exchanger into contact with an external environment at a temperature Text, so that, for each plate Pl^, the nominal value of the density of the heat exchange flow through the exchange surface Scch, , of this plate Plij is equal to qw^, characterized in that, during operation under nominal conditions: - inside each channel, the flow of the heat transfer fluid is laminar, and - for each pass Pa; of the heat exchanger used, the hydraulic diameter Dhi and the surface Ss; of the flow section Si of the channels of this pass are comprised, respectively, between 0.7*Dhi>opt and l.3*Dh i>opt and between 0.7*Ssi>opt and l.3*Ssi>opt, where the hydraulic diameters Dhi>opt and the surfaces Ssi>opt are those calculated for each pass Pa; by implementing a dimensioning method in accordance with any of the preceding claims to size an exchanger intended to operate under the same nominal conditions and having the same structure and by choosing: - Hech and Lech equal to, respectively, the height and the width of the parallelepiped of smaller volume which entirely contains the heat exchanger, and - the volume Vmax equal to the volume of this parallelepiped of smaller volume which entirely contains the heat exchanger.

6. Method according to claim 5, wherein for each pass Pa; of the heat exchanger used, the hydraulic diameter Dh; and the surface SS; of the flow section Si of the channels of this pass are comprised, respectively, between 0.9*Dhi>opt and l,l*Dhi>opt and between 0.9*Ssi>opt and l,l*Ssi>opt.

7. A method according to any one of claims 5 to 6, wherein the heat transfer fluid is water.

8. A method according to any one of claims 5 to 7, wherein, when using the heat exchanger, the inlet is above the outlet and the passes Pa; are arranged one below the other in increasing order of index i so that the heat transfer fluid flows from top to bottom.

9. A method according to any one of claims 5 to 8, wherein the external medium is ambient air.

10. Two-phase heat exchanger for implementing a heat exchange method with a heat transfer fluid according to any one of claims 5 to 9, this heat exchanger comprising: - an inlet orifice (42) through which the heat transfer fluid enters in one of the gaseous and liquid states and an outlet orifice (44) through which the heat transfer fluid leaves in the other of the gaseous and liquid states, - Npa passes Pa,, each pass Pa; comprising Nc; channels, where: - the number Npa is the total number of passes Pa; in the exchanger, this number Npa being greater than two, - the index i is a serial number included in the interval [1; Npa] which identifies the pass Pa;, - the channels of the same pass Pa; all extend parallel to the same longitudinal direction from a respective inlet to a respective outlet, - the length of each channel of the same pass Pa; between its entrance and its output is equal to a length L;, - between their inlet and their outlet, each channel is fluidically isolated from the other channels of the same pass Pa,, - the hydraulic diameter and the flow section of each channel of the same pass Pa; are constant over the entire length L;, - all the channels of the same pass Pa; are practically identical, that is to say that the hydraulic diameters and the surfaces of the flow sections of these channels are included, respectively, between 0.9*Dh; and l,l*Dhj and between 0.9*Ss; and l,l*Ss;, where Dh; and Ss; are, respectively, the average hydraulic diameter and the average surface of the flow section of these practically identical channels, and - in the same pass Pa,, the channels are grouped within Npl; plates Plij which each contain NcLJ channels and, inside each plate Pl^, the Nc^ channels are arranged next to each other in a transverse direction perpendicular to the longitudinal direction, where the index j identifies the channel plate Pf j among all the plates of the pass Pa;, Npl is the number of plates of the pass Pa; and Nc^ is the number of channels in the plate Pf j, - each plate Plij of each pass Pa; has an exchange surface Sechij with an external medium, this exchange surface Sech^ extending mainly in a plane parallel to the longitudinal direction of the pass Pa; and to the transverse direction of this plate Pl^, - the distance Ejj between two immediately consecutive plates Pl^ and PlIJ+i is greater than the hydraulic diameter Dh;, - an inlet manifold Co which fluidically connects the inlets of the channels of the Pai pass to the inlet orifice, - collectors C; which connect in series the different passes Pa; in order of increasing hydraulic diameters when the heat transfer fluid leaves in the gaseous state through the outlet orifice and in order of decreasing hydraulic diameters when the heat transfer fluid leaves in the liquid state through the outlet orifice, so that the heat transfer fluid can circulate from the inlet orifice to the outlet orifice by passing only once through each pass, for this purpose each collector Q being arranged between the previous pass Pa; and the following pass Pai+i, where the previous pass Pa; and the following pass Pai+i are those located, respectively, immediately upstream and immediately downstream of this collector Ci in the direction of circulation of the heat transfer fluid inside this exchanger,

11.

12.

13. - a CNpa outlet manifold which fluidly connects the outlets of all the channels of the PNpa pass to the outlet orifice, this heat exchanger being suitable and intended to operate under the following nominal conditions: - the total mass flow rate of the heat transfer fluid inside the exchanger is equal to mT, - the pressure and temperature of the heat transfer fluid at the inlet are equal, respectively, to Pin and Tin, - the volume enthalpy of the heat transfer fluid at the inlet is equal to hm>in, and - the temperature of the external environment is equal to Text, - for each plate, the nominal value of the density of the heat exchange flow through the exchange surface Scch, , of this plate Plij is equal to qwjj, characterized in that for each pass Pa; of the heat exchanger used, the hydraulic diameter Dh; and the surface Ss; of the flow section Si of the channels of this pass are comprised, respectively, between 0.7*Dhi>opt and l.3*Dhi>opt and between 0.7*Ssi>opt and l.3*Ssi>opt, where the hydraulic diameters Dhi>opt and the surfaces Ssi>opt are those calculated for each pass Pa; by implementing a dimensioning method in accordance with any one of claims 1 to 4 to dimension an exchanger intended to operate under the same nominal conditions and having the same structure and by choosing: - Hech and Lech equal to, respectively, the height and width of the parallelepiped of smallest volume which entirely contains the heat exchanger, and - the volume Vmax equal to the volume of this parallelepiped of smaller volume which entirely contains the heat exchanger. Exchanger according to claim 10, in which the majority of the collectors Ci comprise a single cavity inside which all the outlets of the channels of the pass Pa; and all the inlets of the channels of the pass Pai+i open. Exchanger according to claim 10 or 11, in which at least 80% of the channels of the same pass Pa; are identical. Device comprising: - a heat exchanger (10) according to any one of claims 10 to 12, - a first controllable propellant (14-16) capable of adjusting the flow rate of the external environment between Plij plates of a first set of plates, this first set of plates containing only the Plij plates inside which the heat transfer fluid is in a two-phase state, characterized in that the apparatus comprises: - a first pressure sensor (56) and a first temperature sensor (52), capable of measuring, respectively, the pressure Pout and the temperature Tout of the heat transfer fluid at the outlet orifice of the heat exchanger, and - an electronic control unit (19) configured to: - determine a first saturation temperature of the heat transfer fluid at the outlet orifice as a function of the measured pressure Pout, then - control the first thruster so as to maintain the gap between the first determined saturation temperature and the measured Tout temperature between predetermined thresholds.

14. Apparatus according to claim 13, wherein: - the heat exchanger comprises a collector Ct upstream of which the heat transfer fluid circulating in the pass Pat is a single-phase heat transfer fluid and downstream of which the heat transfer fluid becomes a two-phase heat transfer fluid inside the pass Pat+i, - the device comprises - a second controllable thruster (15-16) capable of adjusting the flow rate of the external medium between plates Pl^ of a second set of plates, this second set of plates containing only the plates Pl^ inside which the heat transfer fluid is in a single-phase state, - a second pressure sensor (54) and a second temperature sensor (50), capable of measuring, respectively, a pressure Pt and a temperature Tt of the heat transfer fluid at the collector Ct of the heat exchanger, and - the electronic control unit (19) is configured to: - determine a second saturation temperature of the heat transfer fluid at the collector level Ct as a function of the measured pressure Pt, then - control the second propellant so as to maintain the difference, between the second determined saturation temperature and the measured temperature Tt, between predetermined thresholds.

Images